How do you compare your data when you move from Universal Analytics to Google Analytics 4 (GA4), or some other analytics package? Learn how to size up a lot of data very quickly using a simple algorithm you learned in high school.
Once you move from Google Analytics to Ga4, you can be assured that your data will not match exactly. In fact it may be off by a percentage.
What is more important is that relative changes from day to day, week to week, and month to month are of the same magnitude in both systems.
Is the GA4 data correlated with the UA data?
For example, if we graphed the Universal Analytics (UA) metric Users against the GA4 metric Active Users, it might look like this:
It is clear that UA consistently reports more Users than GA4 active users on a day-by-day basis.
The blue bars are the users reported by UA and the green bars are the active users reported by GA4 on a daily basis. It’s clear that UA is reporting more users than GA4 is reporting acitve users.
This is to be expected, because active users are calculated differently by GA4 than is users.
What is more important is that they move similarly to each other day after day. In other words, if GA4 is going to report fewer active users, the magnitude of the difference between it and UA should be consistent, day after day.
For most days this appears to be true. But some days, UA reported many more users than GA4 reported active users.
Does this mean that we can’t trust one or the other? There is a way to find out.
Scatter Plots, Not Bar Graphs
The bar graph is a crude tool for comparing two data sets. In fact, any time-series graph is going to disappoint.
What we need is a Scatterplot.
The Scatterplot graph in Google Sheets
A scatterplot ignores the order of the date and instead compares the data on each day. On a day that UA reported 200 users, how many active users did GA4 report? We plot that point.
When we do it for each day in our data set, we might see something like this:
Scatterplot graph of Users from UA vs. Active Users from GA4
What you might notice is that this data lies in a straight line, for the most part. This is a good sign. It means that the GA4 data changes relative to the UA data for each of the days mapped.
This doesn’t mean that it’s accurate, though. Here’s a scatterplot of the same data, but I’ve artificially doubled the daily UA data.
A scatterplot in which the value of one dataset has been doubled. It looks pretty much the same.
This data looks good, but it’s not. How would we know?
Spreadsheets and your high school math teacher give us a simple way to evaluate the data like a boss.
Add a Trendline
First, Google Sheets will calculate a trend line for us. When at science events, we call this a linear regression. This is the straight line that best “fits” the points. If the points look like a line, then the trend line will be a close approximation of the data. In Google Sheets you’ll find this in the Customize tab under Series >.
These features exist in Excel as well.
Check the Trendline box in the Chart Editor of Google Sheets
When we add a trend line to our data, we see this:
The trend line matches the data very closely. But how closely?
That draws a pretty line right along with our data. How closely do the two data sets match? That’s what R2 tells us.
Google Sheets will calculate R2, but this is not enough. We want the equation of the trend line so that we know how closely related the two data sets are.
In Google Sheets, set the Label field to “Use Equation” and check the box labeled “Show R squared”.
There are some mathy looking bits in our legend now.
The R squared value tells us how closely the data “fits” our trend line and the equation describes the trend line in detail.
The R2 number tells us how well the trend line describes our data. A perfect fit would give us an R2 value of one. The closer to one it is, the more likely our two data sets are describing the same thing.
The equation is the one you learned in high school. It’s just the equation of a line.
The Equation of a Trend Line
This is one of those equations that you swore you would never use in math class. Today, it’s going to give you X-ray vision into your data.
y = mx + B
x is the GA4 Active Users
y is the UA Users
The choice of x or y axis is arbitrary for a scatterplot.
m is the “slope” of the line. It’s the “rise over the run”. If we expect our two datasets to be alike, we expect a slope very close to one.
B is the “y intercept”. It is where our line crosse he vertical axis, also called the “y axis” when x is zero.
We’re hoping that our GA4 data is as much like our UA data as possible. If the two were reporting the exact same number each day:
R2 would be 1
The slope (m) of the line would 1
The y intercept (B) would be 0
I compared two identical data sets to show this.
Two identical data sets. The R squared value is 1. The slope is 1. The y-intercept is 0.
So, what if our data isn’t perfect?
If R2 is significantly less than one, the two data sets are not well-correlated to each other. In other words, they are not describing the same thing. If it’s 0.9 or above, we feel pretty good about the comparison. If its below 0.8, we should be worried.
Even if R2 is close to one, the slope (right before “x”) might be significantly less than one. In this case, we would find that that one dataset is adding or subtracting a percentage of the true value. It could be doubling the count of users, or not reporting users on some percentage of the pages of your website.
If the R2 value his close to one and the slope is close to one, we may find the y-intercept to be higher than zero. This means that some consistent value is being added to one or the other dataset. One is counting something that the other is not.
Here are some common scenarios we see in comparing UA and GA4 data, and how the equation would be expected to change.
You’re comparing the wrong data.
Let’s start off by looking at a bad correlation. Here the R2 value and slope are near 0. The y-intercept is very high.
Something is just not right here. Maybe you’re not pulling the data right.
Both the R2 value and slope are near 0. The y-intercept is way above 0. This is data that doesn’t correlate.
Bot traffic is not being filtered in one dataset.
In this example, I’ve artificially added 50 users per day to one of the datasets. This is what it would look like if GA4 was filtering out a consistent traffic source, like bot traffic, but UA was not.
The entire trend line will is lifted by 50 users. Because it’s consistent, the slope and R2 values are not affected. But the y-intercept will rise precariously.
Adding 50 users to one dataset increases the y-intercept, even though the slope and R2 values are near one.
You’re double counting.
It’s remarkably easy to double-count by adding the Google Analytics tag twice. In this case, the slope will be close to 0.5 (or 2.0 if you flip the x and y axis in your scatterplot).
It’s not unusual for us to find a website that is adding pageviews using an on-page tag and a tag manager tag. This will double-count pageviews.
When you double the users reported in one dataset, the slope will approach 0.5 or 2.0.
You are “breaking” sessions.
If you are “breaking sessions” in either dataset, you’ll see inflation of sessions. This will be reflected in the slope. It will be significatnly above or below one.
For example, if you use a utm_ query parameter on a call-to-action button on your site, UA will start a new session, as if the user was just arriving on the site. GA4 doesn’t do this.
If your visitors are going to a third-party site and returning, you can get broken sessions. If you have cross-domain tracking setup in UA but not in GA4, you’ll see something like this for the segement of visitors that visit the other site.
The analytics tag is missing on some pages.
With this example, I’ve added 50% to the dataset on the Y axis. This simulates the scenario in which 33% of the pages on the X-axis dataset don’t have tags.
Note that the R2 value doesn’t change. However, the slope of the line is well below 1. In fact, it’s about 2/3 of a perfect slope.
When adding 50% to one dataset, you find the slope changing even though the R-squared value is the same.
Revenue, Transacations and Segments
This approach can be used to check most of your metrics and segments.
Not only can you evaluate the data you are collecting, you can evaluate your ability to pull data in GA4 that represents the thinking of the UA developers. UA data is pre-processed differently in UA than it is in GA4.
This is a great way to be sure you’re pulling similar data segments.
Compare Google Analytics to your sales data.
If you want to be sure Google Analytics is collecting ecommerce data, you can compare transactions from GA to transactions from your backend, such as Shopify, BigCommerce, Magento, etc. This approach is great for that.
This is one of the first things we do with or new Conversion Catalyst clients.
The graphs look the same. Don’t be fooled.
Be careful when you move from Google Analytics to GA4.
In all of these examples, the scatterplots look pretty much the same visually. However, our high school math teacher has equipped us with the equation we need to diagnose our data.
When CRO and SEO work together, there is a cycle of increasing advantages, according to technical search engine marketer Jason Fisher. Find out how Jason combines SEO and CRO for a one-two punch that delivers results.
What if you were Google Search’s therapist? You would be tasked with understanding the mind of Google’s search algorithm, simultaneously seeking to understand the Google mind, and trying to help it make better decisions about the world it manipulates.
This, as it turns out, is the job of the search engine optimizer, or SEO. Maybe we should call them SETs: Search Engine Therapists. For they must not only understand the search algorithm, but they must also be able to help it make better decisions.
And they must do this with the active resistance of Google. Like humans, Google secretly does not want to be helped.
Our intrepid SETs will never have a complete understanding of the Google search mind. So, we are stuck with a simple mantra: “Take the best search traffic we can get and let the website sort it out.”
I’m the guy that focuses on “sorting out” the traffic. Making a website better at finding the buyers in your traffic is called Conversion Optimization, or CRO.
Here lies the delicate balance between making the search mind better at its job (SEO) and making the most of what comes your way (CRO).
Google is a cantankerous patient who makes the therapist pay for the privilege of helping it. We need to use all our weapons to maximize this traffic source.
This is why I invited Jason Fisher onto Intended Consequences. He is a search therapist who gets that the “sorting out” part of the equation is important.
* * *
Jsason Fisher has been doing search marketing for a long time. He cut his teeth on search engine marketing right at the beginning of the century, when search marketing was the like the Wild West. Some years later, he was working for a leading link development company, where he learned the power of back links.
He is filling the technical SEO skills gap, primarily for agencies. And he’s the kind of guy that throws around phrases like “competitor link graphs”.
I was curious how a guy like this sees conversion optimization in relation to his work. Here’s what I learned.
CRO vs. SEO: Advice to new online businesses.
We do a lot of consultations with businesses that aren’t ready for our conversion optimization services. My advice to them is this:
Start working on your organic search strategy.
Use search ads to begin bringing traffic.
Then look at optimizing the site to maximize revenue.
Organic search optimization takes 12-24 months to implement. It takes time to research, to create valuable assets, and to build relevant backlinks. But this cannot be easily taken from the business.
Paid search can begin working immediately, but is subject to the bidding of your competition and the whims of the search engine ad algorithms.
Of course, you don’t have to put conversion optimization on hold during this time, especially if your paid search campaigns aren’t yet profitable.
What are the components of SEO?
For established websites, Jason discusses four main components of an organic search program:
Accessibility
Indexability
On-page Success Factors
Off-page Success Factors
Portion of a hamburger with label “On-page/Off-page Factors”
These on-page and off-page success factors are the things the search engines look at to determine your authority for certain queries.
Off-page Link Graph: The quality and quantity of links
On-0page meta data: Title, internal and extermanl linking, etc.
Host performance is an on-page factor. If you have a good tech stack, you should be good.
Over-indexing and Under-indexing
An often overlooked aspect of your search strategy is that the wrong pages are being indexed.
Over-indexing and under-indexing are important CRO and SEO factors.
Do you have a lot of poor-quality pages indexed on the search engines? This can be a problem.
Do you have quality pages that aren’t indexed? Of course, this should be addressed.
It may seem counter-intuitive that you should remove some content from the search index. Jason says that there is a limited “crawl allotment”. You may be wasting yours on poor-quality pages by marking them as “no index”.
Where does link development fit in today’s digital marketing practice?
Link building starts with creating valuable assets, things that others want to link to. In this sense, Jason sees link development like public relations: you’re trying to get people with authority to link to your work.
It takes a lot of work. Of those solicitation emails that us publishers get daily, only 1% results in a backlink.
Evergreen Content
Content is the meat of the sandwich.
Evergreen content are those assets that can stand the test of time should be a focus of your efforts.
Update them and relaunch. Always be looking for ways to increase the quality. For example, transcribe videos on that page to text.
Content Blueprint
Start with the top keyword opportunities in your space. For those phrases that you don’t have content for, you can retarget existing content to them, or create new content that addresses them.
You are closing the gap on the most important search phrases for your business.
Your goal is to demonstrate to “the machines” that you are the authority in your space. People are going somewhere to get answers to questions relevant to your business. Why not your website?
Some of these topics may seem counter to your intuition. Why put DIY content on your site when you want to do the work for them? Because it tells the machines that you are the authority for those search phrases.
Can we outsource this content?
It depends on how complex your product or solution is. You may need a subject matter expert to truly develop quality content.
Content that gets visitors to stick around.
Content that gets others to link to it.
How can a website convert more visitors to prospects and customers?
Your website needs to
People overthink product and service pages. Jason likes to focus on those things that differentiate your offering.
Yes, testimonials and customer logos can increase your conversion rate. Yet, selling your brand and it’s values make a big difference.
This is harder than adding logos to your page.
It is especially difficult to do your own writing. We are too close to our businesses to present them in a way that the visitors need. Let an external resource start the page and then review it to ensure it reflects your advantages and brand voice.
All of this is the foundation for a great marketing program.
Search optimization provides the information that can be used across your marketing efforts.
Ads
Email
Converison optimization
Phone scripts
You will learn a great deal about your audience and your industry when you look through the eyes of the search landscape.
What do your prospects want? What are your competitors doing? What is missing from your messaging platform?
The content calendar is a marketing roadmap.
Understanding not only what you need to say, but when you need to release messages are key to success in search marketing and marketing in general.
When testing your website, you need to know when to run high-risk tests versus low-risk tests.
For example, during peak holiday traffic, you can learn much from low-risk testing during this high-traffic timeframe.
Write for the machines or write for the visitor?
Jason says that you always write for the visitor. Your selection of topics may be driven by the search engines’ understanding of your industry, but the content is for those entering the search phrases.
Links
Jason Fisher is an organic search marketing veteran with over 15 years of experience. Jason has helped numerous Digital Marketing Agencies all over the country increase their clients organic search channel sales & revenue by building & executing sound SEO Strategies.
Web design is about communicating with people who have problems that your offering solves. Science is about making sure you’re not designing just for yourself. If you do web design for conversion, science will guarantee that your new design will perform better. Here’s how.
This is Alice. She’s a web designer. Alice is doing a website redesign. The design she completes is based on research and her extensive experience. She is confident that this new design will make more visitors choose her company.
Alice the Web Designer
Alice works with Bob, the web developer and Cindy, the marketing manager to finalize her design, copy and images.
Alice the web designer, Bob the web developer and Cindy the marketing manager.
Their boss, Doug, has faith in the team and thus the design. He loops in Emily from Sales. She thinks it is a big improvement.
Yet, we haven’t asked any customers yet. So they invited some prospective customers in to weigh in on the new design.
Most of them thought the new design was a definite improvement.
The focus group was mostly in favor of the new design.
Frank, who worked in reception was concerned. He loved to shop online, and knew that if a product had a 5-star review but had only 12 reviews, he wouldn’t believe that rating.
So, why was 12 opinions enough to believe that this design is actually an improvement? Five of them were company employees, after all!
How 12 people can get it wrong.
Fortunately, Georgia, the company scientist was also concerned. She knew that the internal group was going to love the new design because they had designed it. And he knew that members of focus groups want to like the new design, because they are eager to please.
They’re human.
Instead of replacing the old design with the new design, Georgia setup an experiment. Half of the visitors to the website saw the old design and half saw the new design. Which one would generate the most revenue? What she found was that there was no difference. As far as the company’s prospective customers were concerned, there was no difference.
How did a talented bunch of people, with the help of real customers come to such an erroneous conclusion?
Let’s assume we have 100 people visiting a website. Half of them are more likely to buy with the old design and half of more likely to buy with the new design. You can see that it’s pretty easy to collect a sample of opinions that don’t tell you what is really going on. Pick the wrong 12 people, and you can either bet on a poor design, or throw out a pretty good design.
Most design teams are biased in favor of a design because they created it with no other opinions involved.
Bigger sample sizes mean fewer design mistakes.
We can never ask EVERYONE if they prefer to see the old design or the new one. We have to ask a subset of the population of visitors. Scientists use the variable “n” for the size of the sample we are going to “ask”.
Here’s the problem from a Design Scientist’s point of view.
One designer? n=1
One designer, one developer, one marketing persion? n=3
Add in an executive and a focus group? n=12
At n=12, we are till making mistakes in our design decisions. How big does n need to be?
In our example, a sample size of n=40 gives us more confidence that we are seeing reality.
It’s harder to make bad decisions with larger sample sizes.
Where do we find these bigger sample sizes?
A design scientist has two tasks:
Increase the sample of people opining on her designs.
Increase the quality of the sample of people opining on her designs.
There are three broad ways of getting more n’s to look at your design options:
AB Testing
Trial and Error
Usability Studies
Usability Testing
A usability test is essentially a giant focus group. Thanks to services like UsabilityHub, you can bring 25, 50, or more people to look at your new designs and tell you which communicates better.
Cons: These panels of people are NOT necessarily customers of your product, so their input is less reliable than trial and error or an AB test.
Trial and Error using analytics
To get the “input” of your actual prospects, it makes sense to launch something. Then you use analytics to see if the change made things worse or better. However, you must be willing to roll your new design back if you find that conversions drop with the new design.
We recommend using an AB testing tool to make changes to the page and then use analytics to determine if the change was an improvement. If it was, make the change permanent. If it wasn’t, try something else.
Cons: If there is a shift in traffic, pricing, promotions, competitors, or anything else, your results can be skewed. For example, if your competition launches a sales at the same time that you launch a new design, it can look like the new design is decreasing your sales.
AB Testing
The AB test is designed to overcome the limitations of the other two approaches. It takes it’s sample from your actual web visitors and it controls for changes in the marketplace. For an explanation of how this works see our intro to AB Testing.
Cons: AB testing is limited by the amount of traffic you are getting and requires some developer support.
Guaranteeing your design will outperform the old design.
If you are able to improve the number of brains involved in your web design for conversion, and can bring the brains of actual prospects, there is no reason you should make small sample size mistakes with your website redesigns.
You can guarantee your new design will improve conversions. If your samples say they’re not better, you don’t use them and try something else.
What is the job of an optimizer? Is it just improving conversion rates? If not, what is the goal of a CRO professional and what are the steps of conversion optimization?
Brian Massey, the Conversion Scientist, shares the steps of conversion optimization. He is the founder of Conversion Sciences, and author of the book “Your Customer Creation Equation”.
If you are new to conversion optimization or if you need to take your website conversion to the next level, Brian’s book is a fantastic foundation. It will help you understand the way an optimizer looks at the world and looks at a website.
What is the Goal of Conversion Optimization or CRO?
Let’s start off by talking about what conversion optimization means to me. I don’t see my job as just improving conversion rates, but getting the most value out of every visitor to your website.
In a lot of situations that might mean a conversion to a sale. But even on an e-commerce site, we might want to connect with visitors who aren’t yet ready to buy, so we try to get them to join an email list. We can get value from them by asking them for their name and email address in exchange for fantastic information, a discount, or something else of value.
In the mobile world, a conversion may look like a phone call. Click-to-call is a powerful way for prospects to take action when away from their phone.
There are a number of things we can do to get value from our visitors, and for us, everything is on the table.
The reason I want to lead the life of a conversion optimizer is because we do wonderful things for online businesses. This is probably why we’re so well-liked.
Benefits of conversion optimization.
Benefits of Being or Having a Conversion Optimizer
First of all, we do increase the revenue from the traffic that you’re getting. The net result of that is that it decreases your acquisition costs, your advertising spend.
If more of those clicks that you’re paying for turn into customers, then you get a positive return on your ad spend.
Having a site with high conversion rates often means people are staying on your site longer. They are buying more often. They are typically visiting more pages. Google understands this and rewards you with higher rankings. If people are staying on your site and not “pogo sticking” back to the search results page, well, your organic ranking will likely go up.
We, of course, increase your growth because we don’t just test what goes on your pages. We can test pricing, we can test bundling, we can test new features — we can test the things that are core to your business.
We’ll know which channels are converting best, which things you’re doing well and which you are not. And you can adjust your marketing spend accordingly.
So, conversion optimizers really are wonderful for a business.
Do you know which path your users are taking?
Logically, we might think that a path through our website should look a certain way, but in truth, the visitors want something different.
It’s our job to understand what that desire path is. For instance, in a park like this, people may be avoiding the paved path because it’s concrete which is harder on your knees when you’re running than dirt. It may not just be because it’s the shortest distance.
We want to understand the visitors’ motivations. That is what I spend my days doing. My whole job is to make sure that I’m not using mental shortcuts to make decisions.
Cognitive bias codex.
The 3 lbs. of gray matter between our ears is just packed full of biases, shortcuts, and stereotypes. These biases, stereotypes and shortcuts, cause us to think we’re doing the right things when we’re making decisions about design, or about our products, or about our pricing.
But in truth, we’re doing it wrong for our users. We take shortcuts. We’re not really connecting with what our users want. This whole method, the steps to conversion optimization, is designed to keep us from relying on our biases to make decisions.
The Importance of the Optimizer
When you’re optimizing, you play a really important role in the design process and in your company. You are the one who is double checking the assumptions that are being made. You are the one making sure that those assumptions are what our visitors and our customers want.
The optimizer plays a key role in the design process and in their company.
The Benefits of Conversion Optimization
An optimizer has many benefits. They save time. There’s nothing that wastes more time than launching a campaign, spending your marketing budget on that campaign and then not having it work. So, by doing a little extra work on the front end, collecting some data, you can make sure that your campaigns are going to be more successful so you don’t have to start over and relaunch them.
The Value of Data (and its many uses)
Data is a great way to deal with what we call helicopter executives, executives who feel that maybe the team isn’t making the right decisions. They feel that they have to come in and review your creative and your campaigns, making changes to what you’re doing. Of course, their assumptions are based on the same biases that anyone’s are.
If you are able to say, “Well, we have some data that says this is the best thing,” then they’re more likely to think, “OK, this makes sense. Go ahead.”
You’ve just removed one cook from the kitchen.
How a conversion optimizer should handle agencies and teams.
Oftentimes you’ll get creative from your agency and think, “Is this really effective creative?” Your agency may present you with options and ask you to choose. As an marketer, your answer should always be “I don’t know, go collect some data to find out which one of these ideas is most likely to be the best choice”. This should be the job of your agency’s experimenter. This is a powerful way to manage teams effectively.
Steps to Conversion Optimization: Gathering Good Competitor Ideas
We like to take ideas from our competitors and from other websites that we like, but we often steal bad ideas. Just because our competitors are using them doesn’t mean they’re working.
An optimizer wants to take ideas and test them before stealing them. At Conversion Sciences, we say, “Steal like a scientist.”
Digital Marketing Careers Require Experimentation
If I haven’t made the point abundantly clear, people with the skills of an optimizer are very valuable. And right now these skills are hard to find and expensive. In a few years, these skills are going to be absolutely required. So, if you don’t have these skills, you’re not going to be able to work in premiere digital marketing roles, in digital product management, or run a business that requires the web to succeed.
Experimenters Can Take More Chances
As a conversion optimizer, you can take more chances because you know how to create experiments that allow you to be more creative.
Experimenters take these really creative ideas that would otherwise sound risky, find a way to collect some data, and then understand whether or not that idea is actually going to improve things. You also avoid implementing a bad idea. We call this a “design insurance”.
You don’t have to always play it safe with your campaigns. You can come up with crazy ideas and experiment before you actually launch and take all of the risk.
Being able to take more risks, a CRO expert can get more leads, more sales, and lower acquisition costs.
And, of course, you get more leads, you get more sales, you lower your acquisition costs, you grow your business.
That’s what most people want from their conversion optimizer. But conversion optimizers are so much more valuable.
My day deals almost exclusively with ideas. Ideas for how to improve a website, ideas for how to improve a customer’s journey, ideas for what kind of content we should be putting on the page, ideas for how we should discount, ideas for how we should lay out a page. Ideas for what we should be doing and advertising. For almost anything that’s going to be seen or experienced by the user, there are ideas for improving it.
How to Find the Right Ideas: good reasons to kill ideas
I’m going to walk you through the process of figuring out which ideas are the right ideas.
When we first start with a client, we go through their website and perform an analysis. This includes an analysis of their existing data. We come up with a very long list of what looks like really good ideas for improving the conversion rate.
Our job is to kill some of those ideas and get them off the list so that we can move on to the ones that are good ideas. In fact, the scientific method that I use on a daily basis is designed around this.
The job of an experimenter is to come up with ideas and then find out why that idea is wrong. When you test a hypothesis, you are actually testing against the null hypothesis trying to prove that idea won’t improve things. If you can’t, despite trying everything, then you’ve got a winner.
So, what are the good reasons to kill ideas? We evaluate ideas based on these criteria. Is there a reason that we should keep this idea on the list?
Reasons to Kill an Idea
1. That’s a lot of work
Some ideas require too much work to test and implement. We might say the website needs to be redesigned. That’s very risky because it changes everything. And so we’ll often just pull that idea out right away.
Ironically, this is the way most website redesign is done. 90% of the market is still redesigning websites this way.
They start by hiring a creative agency or bringing in a creative team. That team does a little research at the beginning of the process, and then they make all sorts of design changes based on that research. Then, they push it all out and hope that they made the right choices.
That’s very risky, and so full-scale redesigns don’t stay on our list very long.
2. It’s too small of an idea
Some ideas just aren’t that impactful. For instance, if we had an idea to change something in the footer of a page, we can tell from our heatmap reports that few visitors are seeing the footer area. We would say that’s too small of an idea. It doesn’t have enough of an impact and we’ll drop it from the list.
Likewise, changing the color of a button or changing the font of our headings are low-impact changes. We tend to just get rid of these ideas.
3. No one is seeing it
There are pages on your site that are important to the customer’s journey, but not a lot of visitors are visiting it.
For example, sometimes FAQ pages can be really important to our visitor’s journey. If we had a hypothesis that said we’re going to change the order of FAQ questions, but we looked in analytics and saw that few visitors were actually visiting the FAQ page, we would say it’s probably not a good thing to test.
On the other hand, few people are seeing the checkout process on an ecommerce website, but those visitors are in the process of buying. In this case, we want to keep checkout ideas at the top of our list.
4. I don’t have any data on it
For each idea on my list, I have to ask myself, “Can I find some data on this idea.” This is the question we ask ourselves over and over and over. If I can’t find data on an idea, or I can’t generate data on that idea, then it’s not a testable idea.
A good example might be things on a website that encourage people to visit a physical store. There are technologies to track this cross-channel behavior, but it’s very expensive technology. Even if we have really good ideas about how to drive more people to the brick and mortar store, we really don’t have a way of collecting success data related to that idea. So, that would be something that we would eliminate because we don’t have the data.
Steps of Conversion Optimization: Gather Existing Data
Let’s talk about sources of data. Once we’ve gone through our list, we’ve got things on it which we think are good ideas. We think they are easy to implement or can be implemented in a reasonable amount of time. We think people are visiting those pages, and we think we can find some data on them.
One of the first places I like to look when I’m building a landing page are the client’s paid search ads.
Steps of Conversion Optimization: Gather Existing Data.
Using this made-up example for the U.S. Mint — that’s the part of the US government that prints money — let’s pretend that they’re offering 50% off dollar bills.
Now, you might think this is a crazy offer, better than anything you have. But the truth is that we all have an amazing offer: a great product or service that’s priced right. It saves time. It saves money. It solves a problem. Yet, you still have trouble converting people. Well, don’t be too discouraged, because the U.S. Mint would have trouble giving a dollar bill away for 50 cents.
Go to your paid search team or your advertising team and ask them for a spreadsheet with the last six months or a year’s history of ads.
How many impressions they generated
How many interactions they generated
How many conversions they generated
Go through the data and look for those ads that had a lot of impressions or more importantly, had a lot of impressions AND conversions.
If we look at the third one in our example above, we see it has an interaction rate of 2.8%. That seems like the highest rate, but it was only 37 interactions. This sample size is a little bit too small for us to have confidence in. I’m more interested in those that have 612 or 943 conversions.
It seems that “50% off dollars for a limited time” has a better conversion rate than “Dollar Bills: Buy one, get one.” When I write copy for my landing page, I’m going to favor language that includes “50% off”.
I would not be as excited about “Discounted Dollar Bills” because it had a 0.3% conversion rate and a high enough number of interactions that we can believe that this data is probably accurate.
You see how I can use our ads to understand which words, which headlines, I should be using in my landing pages and in my copy. It’s on my list.
Social media Ad Performance and Conversion Data
We can do this with social media as well. For instance, if we want to put a video on the landing page or video on our homepage, video ads can help us understand what people are interested in.
Use a tool like Ahrefs or SEMrush to look at our competitors’ ads. Find out what words they’re using and if they have a lot of keywords that they’re using.
Previous Email Campaigns Data
Email campaigns are another great source of data. I looked at the subject lines for emails that Conversion Sciences sends. We send emails for new blog posts all the time. I wanted to know which subjects — which titles of our blog posts — were getting the most clicks. I took six months of data and I ranked it based on their click through rate.
Looking at the top ten, we see “writing”, “copywriting”, “persuasion”, “value proposition”, “persuade people”, and “business taglines”. My audience is interested in the words that influence conversions.
I was a little surprised, but we were able to use this data to produce a free copywriting report on how to write copy for conversion writing, and this converts very well for us.
Download these 21 quick and easy CRO copywriting hacks.
In this case, the data really did point us in the right direction. The data created a hypothesis, an idea, and then gave us the data that said you should launch this. Then we used the conversion data, the number of leads that we’re generating on our website, as the final proof that this was a good idea.
Steps of Conversion Optimization: User Testing
Another step in conversion optimization is user testing. Everybody thinks that a conversion optimizer spends most of their time split testing. This is the best data we can generate, the best tools that we can use. But in truth, I want to gather data faster and doesn’t require me to use precious visitor traffic.
We only want the most important and best ideas to go to AB testing while using user testing to figure out which of my creative ideas is best.
Our user testing includes things like a 5-second study. A 5-second study works great when I have three or four different headlines and three or four different images that I want to consider for a landing page.
We’ll use a service like UsabilityHub or Helio and we’ll ask for 25 people to come and look at each of our mockups. The 5-second test works like this: test subjects get to see the mockup for five seconds and then it disappears.
But five seconds in the human brain is quite a long time. After the five seconds is up, we’ll ask questions like,
Does this business seem credible?
What do you think this business does?
Do you know what we were asking you to do?
Where would you click if you were going to take action?
Can you remember any of the bullets or any specific information on the page?
We can score these twenty five people in each of these areas. The image and headline combination that scored the best tells us that it’s probably the best idea.
We now have some data from real world people that is telling us which idea to take to an AB test. There might be a couple of these that score well. So, we want to take the two best ideas to an AB test, but it also means we don’t have to test the others and waste traffic on those.
There are a number of tests that you can use for user testing. Usability hub or Helio, offer a question test where the visitor gets to look at the page as long as they want and answer questions.
A first click test measures how quickly someone can find where they’re supposed to click based on the prompt that you give them. How many of them get it right in test layout or how clear the call to action is on your page.
User testing tools like UserZoom or UserTesting.com allow us to set up a scenario and ask the visitor, for instance, to go through and purchase on a website. We watch them as they try to complete the task. We see where they get confused, where they get tripped up. They’re talking out loud as they’re going through it.
You’re going to see issues in these user tests that you wouldn’t catch otherwise. It can be very enlightening. We can really learn quite a bit from that user testing videos.
More Data Sources: User Intelligence Tools & Reports
Another step of conversion optimization is to look through user intelligence reports. User intelligence is different from user testing.
User testing uses strangers and pretenders. These are people who aren’t actually trying to solve a problem, but we’re using them as a focus group to play with our creative and see how effective it is at communicating with human beings.
User intelligence tools are actually watching your visitors as they interact with your website.
Analytics has the most obvious user intelligence data. Google Analytics is a great behavioral database. It’s all the people who are coming to your website to try to solve a problem. It shows you where they landed, what channel brought them, what pages they visited, how long they were there, where they left, if they bought, how many of them bought, what their computer setup was, what browser they’re on — all of this information is in Google Analytics.
It’s a great database for asking questions. I probably spend at least 10 to 20 percent of every day in analytics, and if I’m working on an analysis, I’ll spend the entire day in analytics, it’s such a rich source of data.
The other thing we use is what are called heatmap reports. They tell us how far the visitors are scrolling when they visit a page, where their mouse is moving on a page, and where they’re clicking. These are great tools for answering specific questions about a page.
You don’t have to be a Ph.D. in science to understand them. If you can read a weather radar map, you can read a heat map.
Here is an example.
Heatmaps of a website page for golf resort in Hawaii.
This is a resort in Hawaii, a golf resort. They assumed that since it’s a golf resort, people who are considering booking a room are going to be interested in golf.
On this page, which lists all their specials, they listed the golf specials at the top. When we went in and looked at where people were clicking on this page, we found out that “Free Breakfast” was most clicked item, even though free breakfast is down near the bottom of the page.
People don’t behave the way you think. What is the cost of breakfast? At this resort, it might be 40 or 50 bucks. If you’re going to save a couple of hundred dollars on golf, it would seem to be a better value, right? Not according to the visitors.
These are the sorts of insights that conversion optimizers love to find.
I also spend time watching session recordings. With session recordings you get to watch visitors as they’re working through your site. You see where their mouse is moving and what they’re clicking on. It takes a while, but you find things that you wouldn’t otherwise discover.
Session recording of golf rates page.
If you watch a bunch of these, you begin to understand what’s bothering your visitors. If I’ve got a specific idea that I’m trying to remove from the list, I’ll spend some time watching session recordings.
Sticky heatmap.
A more advanced conversion optimization strategy is running an eye tracking study. Now, this doesn’t work directly with your website, but you can bring people from your website if they’re willing to take a look. And it’s just amazing that this technology exists because eye-tracking studies used to be so hard.
Submit a mockup to a company called Tobii, and they’ll bring 25, 50, 100 people to look at it. They’ll record what the visitors see on the page using laptop cameras. Laptop cameras have such high resolution that we can tell where people’s eyes are looking on the screen.
We can see what people linger on, what ideas they like, what offers they like, and where there are images that stop them on the page. This information is really valuable if you’re trying to critique your page layout.
Gamification: AB Testing
The last thing that I spend time on is AB testing. Because if we’re going to take something to an AB test, as it’s the best data we can collect, we only want to take the best ideas. And it takes quite a bit of work to get AB test results.
Here is an example of one that we did. We worked with a company called Automatic and they had a plug that plugs into your car and connects your phone to your car’s computer. They came out with this new Pro version, but everybody was buying the Lite version.
Why wouldn’t people buy the Pro version? Sure, it’s more expensive, but it’s so much better. Maybe we’re not communicating how much better it is effectively.
We did a “Thank you” page popup survey asking, “What made you choose Lite instead of Pro?” We found out that people didn’t understand the value of the Pro features.
We created a version of the product page was simpler. It was a shorter list of features, and we only highlighted the things that were most different. This is something you should consider any time you’re offering multiple plans or products on a pricing page.
We designed an AB test. One half of the visitors saw the original page, which we call the Control. The other half saw our variation. The result was a 13% increase in conversion rate for our variation. We also saw an increase in revenue per visit because more people were buying the Pro version.
A/B Testing: automatic pro vs lite.
This achieved exactly what we wanted. The data we collected during the AB test was very reliable, because these tests are designed to eliminate as much randomness as possible.
What are the Steps of Conversion Optimization Summary
If you are going to be a successful digital marketer, you are going to be an experimenter. Your ability to use the tools and data of the trade will determine your future in a data-driven marketing economy.
Can you send a daily email to a business-to-business email list? How often can I email my B2B list? Check out these 4 lessons learned.
One of my favorite conversion strategies is the second chance. The second chance only comes when I have a way to continue the conversation; to get someone to come back again and let me make my case again.
There is no better second chance channel than email.
When entrusted with an email address, and permission to continue the conversation, I have one, two, three or more chances to persuade a prospect to reconsider.
In a business-to-business situation — the considered purchase — in which a decision will be made over a period of weeks or months, email is a true friend. And if it is executed with respect, it is a friend to those struggling with a purchase decision.
The question is, how many second chances am I going to take?
Five Emails an Hour
I tell companies that they can send email as often as their content allows them.
I once got five emails from American Airlines within the space of an hour. Did I unsubscribe? Did I feel spammed? The emails were telling me the status of a flight I was booked on as its departure time and gate changed. The emails were completely relevant to my situation, and were welcome.
If we were to stand by our statement that businesses can send as often as their emails’ relevance allows, we need to understand the dynamics of a high-frequency email campaign.
How Often Can I Email my B2B List: An Email a Day Experiment
The goal of this experiment was to examine the following hypotheses:
Sending frequent email would significantly increase my conversion rate.
Sending frequently would cause an unacceptable number of my subscribers to unsubscribe.
Sending frequent email would reduce my ability to deliver email due to spam reports.
a. The List
We chose a selection of 2000 names from my house list. This list consists of contacts made through personal interactions, meetings and consultations. It is primarily a business-to-business email list.
I would call the list a “semi-warm” list having received email from me only quarterly. This list had received emails on January 11 and April 30. The experiment began September 7.
Your list could easily be generated from social media traffic or search engine traffic.
b. The Content
Because of the frequent nature of these emails, it was important that they provide some value and be entertaining. This proved to be a significant challenge.
Link to relevant content online or registration for a live event
Offers varied, including an invitation to subscribe to my mailing list, registration for a live workshop and an invitation to a Webinar on writing for landing pages.
Subject lines included “Are you the victim of the Email Invisibility Ray?,” “Social Media: Marketing from my La-Z-Boy,” and “Why eight-year-olds beat me at Chess.”
3. The Frequency
Emails were sent daily, Tuesday through Friday for two consecutive weeks. Eight emails we sent in all.
High Frequency Email Campaign Test Results
1. Email Performance vs. Social Media
We’ve had relatively good luck using social media to drive traffic to my site. However, in Figure 1, you can see that the email resulted in significant increases in traffic, even outperforming our summer social media experiment.
Figure 1 • Traffic sources overview: email effect on site traffic.
Hypothesis: “Sending email would outperform social media marketing.” True
One interesting note is the rise in search engine traffic at the time of the email. This underscores that click-through rate is only a partial measurement of email effectiveness.
2. Increased Conversion Rate
It is probably not surprising that sending email to a targeted list is going to result in more conversions. However, keep in mind that my social media networks are also quite well-targeted.
As expected, both conversions and conversion rates for new subscribers increased. We can also attribute thirteen (13) workshop registrations to this email series, generating almost $1300 in sales.
Just looking at new email subscribers, the conversion rate for our social media experiment were 2.5%. For the period of this email, conversion rates were 7.6%.
Figure 2 • Emails’ Effect on Conversion Rate.
Hypothesis: “Sending frequent email would significantly increase my conversion rate.” True
3. Opt-out Rates
This was the metric I was most interested in examining. How would unsubscribe rates change over the course of the experiment?
Figure 3 • Open rate, Click-through rate and Bounce Rate for each drop.
I consider an unsubscribe rate of 1% or less acceptable and expected in any email that asks the reader to take action. So, I got pretty nervous as unsubscribe rates rose to 3.2%, and stayed well above 1%. Over the course of the experiment, 15% of the list unsubscribed.
There are two ways to look at this:
We lost 15% of our prospects.
We identified the 85% of list members that are interested and qualified.
If my goal with this list was primarily to sell, I would consider the 15% loss to be acceptable and even desirable. This is called Shaping your list.
However, my goal is to evangelize conversion and to educate, so the opt-outs represents a pretty significant loss of reach.
From a brand perspective, there were very few negative comments, and many positive ones.
Given the opt-out rates, would I do this again. The answer is a resounding yes.
Hypothesis: “Sending frequently would cause an unacceptable number of my subscribers to unsubscribe.” False
4. The Effect on Deliverability
The other negative effect that frequent emails can have is an increase in spam reports.
For most service providers, deliverability is the inverse of the bounce rate. If my emails are reported as spam, we would see an increase in bounces. Intuitively, when shaping a list, you expect bounce rates to drop quickly as bouncing addresses are removed from the list.
For our experiment, the bounce rate began at 2.5% but quickly dropped, leveling at an imperceptible 0.06%.
One reader was kind enough to let me know that they had “spammed” my email. I used the site MXToolbox.com to see if my domain had been placed on any black lists. However, it would be our Email Service Provider (ESP) that took the hit if spam was reported. This is one big value of an ESP. They keep themselves – and you – off of black lists.
Figure 4 • Unsubscribe Rates for the Email Series.
Another measure of reader interest is open rates.
Email service providers count the number of times a special image is downloaded to establish open rates. Since many people have images turned off in their email client, the open rate is not an accurate measure of actual opens.
However, I would interpret a steady drop in open rates as a sign that the list is becoming fatigued with my communications. Open rate can also be a good indicator of the quality of your subject line.
Open rates were relatively flat, dropping on Fridays.
Overall, I believe that few of my readers reported these emails as spam.
I attribute this positive outcome to the non-promotional nature of the copy, even though the emails were clearly promoting our email list, workshop and webinar.
Hypothesis: “Sending frequent email would reduce my ability to deliver email due to spam reports.” False
How Often Can I Email my B2B List Conclusions
With some simple analytics in place, we can pretty easily establish the ideal frequency of our email campaigns. Based on these results, we should be sending email more frequently. You will probably come to the same result. However, we tested a certain kind of email with this experiment; an email that is informational and entertaining as well as promotional. This style of email requires a bit more work and creativity on our part.
The payoff is quite clear.
Email is a more effective channel in a B2B sale than is social media. It is also a great way to get more out of your search engine and advertising traffic. When you get an email address, you get a second chance at the sale. And a third, fourth and fifth chance.
For the complete content of the emails sent during this experiment, and the results of some split tests conducted, visit.
Interested in setting up your own conversion marketing laboratory? Run your own secret science experiments? Brian Massey, the Conversion Scientist, will tell you how.
Sometimes it’s better to ask forgiveness than permission
Warning: this information will make you a more successful marketer, but may also put your immediate job in jeopardy.
To be a true hero, you must have two things:
An arch nemesis
A secret
Unfortunately for those of us in marketing, our nemesis is often the organization in which we work; that Dilbert inspired, plodding structure full of people that think they know how to market. Such a beast is often resistant to our most powerful weapons, such as positive results.
The best way to defeat such a daunting foe is through patience and stealth. As marketers, we must build our strength, our knowledge and our skills.
How to set up your conversion marketing laboratory.
Your Secret Conversion Marketing Laboratory
I propose that you consider building your own secret conversion marketing laboratory, your own Xanadu. This is the place you go to explore new marketing strategies and ask questions that others may not have the guts to ask.
Questions like:
What if we used more copy on our landing pages?
What if we tried an interesting headline?
Would audio or video increase our conversion rates?
Will social media work in our business?
These are the questions that take time to sell internally, especially when you don’t have the data. These are the concepts that IT is designed to thwart. It’s time to unshackle yourself. Build your own conversion laboratory.
Rules of Engagement
Now, as heroes, we want to do good in the world. This means doing no harm to our organization’s brand. We don’t want to work against our organizations already plodding attempts to communicate.
We want to minimize cost – most of us aren’t Bruce Wayne – and maximize automation. This will make our time in the lab most productive.
I cover all of the guidelines in my Search Engine Land column Setting Up Your Own Conversion Lab, Part 1.
Why Do We Need A Conversion Marketing Laboratory?
Because conversion marketing is a momentum game. It requires trying things to find out what works best. It requires rapid question-test-analyze-question cycles. And sometimes we have to test unintuitive assumptions to understand our audience.
Without the lab, there are blocks to momentum.
IT has their gatekeepers that slow our testing cycles. Management wonders why we aren’t writing a press release or blog post.
While most marketing departments think they know best, our lab lets our visitors tell us what they want. This is powerful knowledge. There are some big wins to be found in the lab, especially at the beginning.
The Secret Conversion Laboratory
Your secret conversion lab should be set up with a few best practices to be successful.
Consistent measurement trumps accurate measurement. Conversion marketing means making decisions based on data. Analytics provide that data.
We aren’t interested in an analytics implementation that is accurate down to the visitor. Instead, we want analytics that are sufficiently correlated to reality.
This is scientist-speak for “when things change, our measurement changes by about the same amount.” When more people visit, our metric “visits” goes up by about the same percentage. It mirrors reality.
Don’t waste your precious time trying to get accuracy in measurement. Good enough is good enough.
Most analytics systems are easy to set up, or are competently integrated into most of the online services you’ll be using in your lab.
Equipment cost must be “under the expense line”. The secret lab is, by design, not going to be a budget line item. That defeats the purpose.
Instead, you need to select tools that are free or cheap enough to purchase and implement without going through the budget process. They need to be expensible.
Avoid IT obstacles. The equipment you use in your conversion lab must not require IT resources to set up and use. IT is too often a bottleneck.
It should be highly automated. We must get our marketing duties done with excellence, so our conversion lab can’t take a large chunk of our precious time. If you’re off in the lab for hours at a time, people will begin to wonder. It draws attention.
We will be looking for tools that automate the lab, and solutions that collect and aggregate data for us.
Your efforts should not harm the live web site. Our goal is to become better at marketing for our companies. As such, we should do no harm. Our lab should not:
Violate company brand guidelines
Compete with corporate sites on the search engines
Take significant financial chances
Violate compliance requirements in regulated industries
Circumvent or disregard your company’s privacy and permission policies
Basically, we want to do small tests, learning things we can use to help the company sell more and dominate online.
Beakers, Bunsen Burners and Mass Spectrometers
We are fortunate to have many of the tools needed in our lab available for free or at low cost.
You will need tools to:
Create and host content of many types.
Put measurement equipment in place
Heat up your experiments with traffic sources
Select the right content management system to host your experiments
Cape and tights are not required
It may be tempting the done a hero’s uniform once you begin to feel the power of what you learn in your lab. Honestly, It’s best to stay under the radar.
Let us know which tools you find in your lab in the comments, and please share any interesting results you get from your experiments.
Discover how AI marketing tools truly work and find the answer to the question: Can they really increase your website’s conversion rates?
Do you know how machine learning is impacting conversion rate optimization for marketers? We all know what the acronym “AI” stands for: “As If”. Data scientists are telling us that by using AI, they’ll will be able to create a predictive model of the visitors to your website that will tell you exactly who is ready to buy.
I say, “As if.”
We may marvel that such things can be done, but we also recognize that these things require a great deal of data and the skills of some serious brainiacs to get a machine to tell us something we don’t already know.
The truth is, you are probably already using “AI”, or more accurately, machine learning in your marketing. It’s hiding in the tools we use, like monsters under our bed. Machine learning and the more sciencey-sounding AI will change the way you take products to market, but your human mind will still be needed and loved.
Things like AI-driven predictive models are exciting, because our job as marketers is to predict the future. We’re like that exotic fortune teller gazing into an empty tea cup or a crystal ball.
We say things like, “If you give me a budget, I’ll generate six times that amount in revenue.” This is is like saying, “If you put a chicken foot under your pillow, you will find true love.” As if.
But this is what we do, and the data on which we base our predictions is often no more valid than the layout of tea leaves at the bottom of a cup. Our brains are wired to find patterns in anything, even when a pattern isn’t really there.
If I came into your office and said, “The last three leads we generated were all visiting the website using a Firefox browser,” your brain would jump to the conclusion, “If I can get more Firefox users to visit our website, I’ll generate tons of leads.”
Do AI marketing tools impact your website conversion rates?
Purveyors of AI, or more accurately Machine Learning (ML), would tell us that the machine doesn’t make mistakes like this. Our 100% genuine intelligence just doesn’t stack up to their Artificial Intelligence.
The problem is that machines will make exactly the same mistake if we don’t give them lots of data.
Just as machines need data, we know that we need more data before we start an ad campaign targeting Firefox users. We’ll ask our analytics person to pull together all website visits for the last year, and calculate the conversion rate for each. This increases the size of our dataset from three to many.
If this analysis goes the way of most analyses, we’ll find that there’s not a meaningful difference in conversion rates among browsers. Most experiments end up being inconclusive. That’s just the way it is.
In this scenario, we “wasted” an hour of our data scientist’s time, an hour of our time, and another twenty minutes explaining to our boss why we were so unproductive today.
“What if,” the AI crowd says, “you could get a machine to sort through your data looking for clues and figuring out who’s more likely buy. You don’t have to waste your time. Let the machine do it.”
This is an exciting proposition. The machine wouldn’t just look at the browser. It would look at the time of day, day of week, and week of the year that visitors converted. It could consider the device being used, screen size and operating system. It could add in the source of the visit, the number of times a visitor has been to the website, and whether the visitor has bought before.
After crunching through all of your analytics data, the machine would give you a percentage chance that the next visitor to your website will convert. And here comes a person with a Safari browser on a Mac computer at 3:30pm EST on a warm Tuesday afternoon who’s never been to the site before.
The machine might spit out, “There is a 51% chance this person will complete the lead form.” Actually, the machine will just say, “0.51”. Machines are so boring.
It’s amazing that a machine can so accurately predict a human being’s behavior. This is incredible.
But, is 51% good? And if this is true, what should my website do differently to make this Safari visitor more likely to buy? Do I reduce the price by 49%? Do I flatter this visitor for being above average? Do I ignore them?
This is “the rub” with machine learning. The machine can’t tell us what to do with the data it gives us. There are systems that will tell us if a visitor is “at the top of the funnel” or “in the consideration phase.” Still, what do we do with that? A price-sensitive buyer may want to see a discount when “at the top” of their purchase process. A relational buyer may not care about discounts until they’re “at the bottom,” ready to buy.
The machine won’t tell us, “Target Internet Explorer visitors coming late at night on a Windows computer during the springtime months with a picture of a cat.” It spits out the probability for each visit: “0.51, 0.34, 0.71, 0.92”.
Wait! A 92% probability? Is that important!? Well, no. They’re probably going to buy no matter what we do. “As if.”
AI-Driven Results
Scoring customers in a customer relationship management (CRM) platform has required that marketers hand-code the algorithm. We decide which actions indicate that a prospect is moving closer to buying. We decide how to value each action. It can work, but it isn’t rocket science – or AI.
Alternatively, we can dump sales data into a machine learning algorithm and let it calculate the probability that each prospect will turn into a customer. The sales force can focus on those high-probability clients and disregard the low-scoring leads. It’s using past performance to predict the future, and should be more accurate than arbitrary assignment of values to actions.
This is how machine learning is entering your life as a marketer.
AI Conversion Rate Optimization: Can AI Marketing Tools Increase Your Website’s Conversion Rates?
Amazon famously introduced product suggestions to the eCommerce world. “People who bought this also bought that and that and that.”
It’s not an easy problem to solve. There are a lot of variables to crunch and it has to be done quickly. This is a prime area for AI.
Mailchimp launched a similar tool to add product suggestions to the emails of its eCommerce clients. Every time you send an email to someone, Mailchimp will include a few product suggestions at the bottom of the email. The machine learning algorithm will compute the probability that one or more products will appeal to a subscriber, based on the behavior of all email recipients. Those products with the highest probability get added to the email. This prompts the visitor to buy.
As if.
It’s hard to know how well the machine has learned what your visitors buy collectively. This is the limitation of AI. We can’t really see what is inside the box. All we get is a number.
If you implement a suggestion engine on your website, we recommend running an A/B test to measure its effectiveness. This is done by adding the “Also bought” suggestions for half the visitors and hiding it for the other half. This will give us some conclusive data about how the suggestion AI is performing. Is it increasing the order size on average, or reducing it?
Seven risk reversal tactics to increase B2B lead generation and the quality of your prospect list you can steal.
Even in B2B lead generation, you can “reverse” the perceived risk of completing a form. Risk reversal tactics are an important part of conversion rate optimization.
There are lots of reasons that someone wouldn’t fill out your lead form, even if your white paper, podcast or webinar are free.
“I’m going to get a sales call”
“I’m going to get a bunch of spam”
“I may not be able to join the Webinar”
“They’re asking for too much information”
“I don’t really know this company”
Well, even in business-to-business lead generation, you can “reverse” the perceived risk of completing a form.
“Hey, you. Yes, you. Come here. You wanna white paper? Well, I have a deal for you. All you gotta do is give me your contact info. Hey, you got nothin’ to worry about. I’ll take real good care of the information. Honest.”
I’ve been studying the landing page techniques of what I call “The Bad Boys of Conversion.” These marketers, often called “affiliate marketers” or “infopreneurs” get high conversion rates for everything from get-rich-quick systems to health supplements.
My goal is to understand how the well-tested tactics they use could be applied to more conservative business sites.
One of the most important techniques used by these marketers is “risk reversal.” Money-back guarantees are perhaps the most common form of risk reversal. Are these tactics useful for business-to-business lead generation?
Absolutely, but a different kind of risk reversal is in order.
Risk Reversal Tactics do work on Business professionals
Business professionals respond to the same incentives that consumers do. However, consumers are risking their money, while businesspeople may be risking their reputation, and their decisions could affect their jobs. This makes business prospects more risk averse.
Thus, using risk reversal strategies becomes even more important in a B2B transaction.
You can significantly increase both your conversion rates and the quality of your prospect list by using some of these risk reversal tactics on your lead generation pages.
7 Risk Reversal Tactics For B2B Lead Generation
Remove the risk up front
Make your white paper, webinar, or video free. Before you say, “Lead generation content is always free,” keep in mind that your prospect is “paying” for the content with their contact information. To them, it isn’t exactly free, and the decision to give you their contact information is similar to the decision to spend money.
Here are some ways to remove risk up front.
1.Don’t ask for contact information.
If you have a piece of content that really nails your value and generates qualified calls give the content away. There is no risk to the prospect, and broader readership can be expected.
Removing risky fields can also lower the “cost.” Fields that increase the cost of lead generation content include address, phone and many qualifying fields. Do you need their address? Do you need to know their marketing budget, really? These risky fields can decrease your conversion rates.
Ask for only the information that you need to qualify and contact them, and eliminate all the rest.
2.Make your forms optional
This allows the prospect to select the level of risk they are willing to accept.
Completed forms will indicate better qualified prospects, prospects that already have a level of trust with your brand. Partially completed forms, less so. Empty form submissions indicate prospects that you may not want to waste time on anyway.
3.Emphasize respect of privacy
On your landing page, best place to put risk reversal is near the submit button (but it shouldn’t ever say “submit”).
The simple phrase, “We respect your privacy” communicates two things: first, that you have a privacy policy, and second, that privacy is important, and you’re not likely to abuse this information, or give it to inbox-clogging spammers.
It is common to make the word “privacy” a link to your privacy policy, but that may not be wise. In his excellent book “Always Be Testing: The Complete Guide to Google website Optimizer”, Bryan Eisenberg recommends testing this. The link takes an interested visitor away from the landing page, and this may decrease conversion rates.
4.Ask for opt-in, not opt-out
If you’re asking a prospect to sign up for email communication, you should explicitly ask for permission. You may feel that this is implied when someone provides their information to you, but any uncertainty increases risk.
Tell the prospect can they can expect and let them opt-in by checking a box. This is an opportunity to sell your email communication.
“Check this box for occasional tips and best practices by email on widget use.”
or
“Subscribe to “Widget News,” the authoritative email newsletter on widgets and their application.”
Allow opt-out at any time. Let prospects who are signing up for email communication know that you will be offering opportunities to opt-out of the email with each communication.
“You can unsubscribe at any time.”
You can use any specifics from your privacy policy, such as “we do not tolerate spam,” and “Your information is digitally encrypted.”
5.Spell it out
Transparency is a great risk reverser. Tell them on the page what will happen when they click the “Get your report” button.
“You will be taken to a page that contains a link to the report. You will also receive an email asking you to confirm your email address. We don’t tolerate spam. Based on your input, a respectful sales person may call, and this is your opportunity to ask any questions about widget use in your organization.”
This risk reversal tactic will lower your conversion rates, but should result in a better-qualified list.
You should also tell them what they won’t get. “No one will call” is a common risk reversal phrase. Of course, this limits what you can do with your house list, and you should be sure to adhere to any limitations you promise.
6.Add risk and then remove it
On the scales of a decision you have risk balanced against value. If your landing page does a good job of building the value of your offer, you will find prospects more willing to take risk, and complete the form.
One great way to communicate value is by setting a price. Consider charging for your content. Then offer a money-back guarantee to reverse risk. No questions asked.
Does this sound like raising the prices in a store just so the owner can run a sale? Yes it does. But our goal is to communicate value—not to boost our profit.
This strategy will reduce conversion rates, but it will deliver highly-qualified prospects.
For example, charging for a product trial—a common lead generation offer—will often ultimately deliver more paying customers than a free trial, even though it reduces the number of people who try the product. Since we’re not really trying to make money on the offer, we won’t be concerned about those who request a refund just to get free content. We should all have content worth stealing.
7.Always deliver on your promises
These risk reversal tactics have two edges. If you fail to deliver on your risk reversal promises, you are damaging your brand and losing potential customers.
You must be able to support your risk reversal claims. You must enforce your privacy policy internally. Don’t email prospects if they opt-out, and don’t let a salesperson call someone on your list if you promised you wouldn’t.
Above all, deliver the content that they “bought” with their contact info and their attention.
I use a unique email address for every site I provide personal information. When I get spam, I check the address to which the email was sent, and that tells me who’s been stealing my contact information. With services such as OtherInBox, more professionals are using this technique. It’s more and more likely that you will get caught, and your brand will suffer.
You don’t want to join the “bad boy” marketers, do you?
In general, you can reverse risk by
Telling prospect how you will treat their data
Telling prospects what to expect if they complete the form
We share 7 proven strategies to improve your email open rates. Try them out and give your email content a chance to convert.
Email is one of the highest converting marketing channels and few activities drive a higher ROI than building an email list. But having 50,000 email subscribers doesn’t really matter if you can’t get any of them to open your emails.
We’ll show you seven relatively simple (yet criminally underused) strategies that are GUARANTEED to improve your email open rates. By the end of this article, you will have an actionable plan for significantly boosting the ROI of every email you send.
7 Surefire Ways to Improve Your Email Open Rates.
1. The Best Way to Improve Your Email Open Rate: Work on Your Subject Lines
Your email is desperately fighting for attention in your prospect’s inbox. Your job is to differentiate yourself within that inbox, and the only way to do that is with your email subject line. But it’s not just about standing out. What is most important to boost those email open rates is that your subject line is relevant to your audience.
Let’s take a look at some examples from my inbox:
Work on your subject lines if you want to improve your email open rates.
Sujan’s email uses a pretty standard article headline. This will really only appeal to me if I want to get more out of my video, and that’s actually a good thing. Using a good blog headline is usually a fine strategy for email subject lines.
Brian’s email will appeal to me if I feel confused or have been searching for some specific answers. That said, this subject line won’t really be enough unless I already have built some trust with the sender. In this case, Brian has already done a great job of establishing himself as my go-to SEO expert, so it works.
Mary’s subject line makes use of an eye-catching emoji, which is good for differentiation. At the same time, the promise is a bit too broad for someone like me to be interested. In all likelihood, however, Mary is targeting a less experienced audience, which might be intrigued by this subject line.
Drew’s subject line is a bit confusing, but it also provokes my curiosity. He is hooking me in with this idea of “9/10 experts DISAGREE”. That’s interesting, and now I’m curious to see what they disagree about.
Remember that the goal of a subject line is not to get just anyone to open the email. You have a specific audience you’ve built, and it’s important to use the types of headlines that will resonate with that audience.
2. Match Your Lead Magnet To Your Content To Your Product
There’s a lot of content out there about how to build an email list. But one of the biggest mistakes I see businesses and bloggers make is thinking too short-term with their lead magnets and other lead acquisition strategies.
Email marketing is a scalable form of relationship marketing. It’s meant to be a long term pursuit, and accordingly, it’s important that what you are talking about remains consistent throughout that process.
If you attract someone to your list via an ebook on increasing web traffic, then start emailing them about how to run an ecommerce store, and then try to sell them a product on increasing productivity, you won’t have much success.
Every part of your sales funnel should be designed to appeal to the same customer profile.
When someone signs up to your list to hear about a given topic, the emails they receive should also be on that topic. By aligning your lead magnets, emails, and products, you can significantly improve your conversion rates across the board.
3. Split Test Until You Find The Optimal Send Time
One of the easiest ways to increase your open rate is to send your emails at the right time.
How do you find the right time? Two words: AB testing
This requires a little bit of intentionality upfront, but it’s so easy to do, and it can have a MASSIVE impact on the reach of your content.
Here’s one way to do it.
Schedule your next 7 emails to go out on different days of the week. For each email, break your subscribers into 4 identical groups and send to each group at a different time of the day.
This will give you a total of 28 different send times to compare against each other. If you have a smaller list, you might need to run this experiment several times in order to get statistically significant results, but what you should start to see is that certain days and times will consistently perform better than others.
By sending at these optimal times, you can significantly boost your open rates.
4. Delete Prospects that Never Open your Emails
Sometimes email subscribers just have to go. It’s hard to part with those emails you spent so much effort collecting but evidently they are not responding to your standard email campaigns. Cleanse your email list regularly as it will also help you improve deliverability. Gmail and Yahoo emails will pick up on this and send your email to their junk folder where nobody can see it, much less open it. And this makes a huge difference to your email open rates!
If you don’t want to get rid of them, split your audience and send them a very different kind of email. A farewell email. Help them move on or open your email to stay.
5. Deliverability is Key to Improve Your Email Open Rates: Resend To Unopens
This is the easiest strategy we’ll discuss today. Even when you do everything right, a lot of your emails will never even be seen.
Thanks to Gmail’s inbox categorization. A lot of emails now fall under the promos tab. There are some things you can do to try and get in the primary inbox, but that’s a bit too complicated for today’s post, so instead, I’m sharing a simple trick that anyone can use to instantly increase opens, often by an additional 25%.
Simply resend the email to anyone who didn’t open it up the first time. I like to wait around 3 days to resend, but you can experiment for yourself and see what happens.
One email is nice, but two could get the job done.
In the example above, the original email performed pretty well, so the resend was a bit less powerful than usual. But even if your list is as small as 5,000 subscribers, that additional 7% in total opens means nearly 400 more people are seeing your message.
6. Be Authentic: The Email Campaign Subscribers Look Forward to Opening
One of the first things people notice when signing up to a new email list is tone.
A lot of marketers and business owners nowadays are just running email marketing because they’ve been told to. They are following a template and trying to sell products. And while there is nothing wrong with that, it’s very, VERY obvious to your readers.
The most successful email marketing campaigns often come from the biggest personalities. When you sign up for their list, you aren’t introduced to a sales pitch. You are introduced to a personality.
You read personal stories about their background and experiences. You read about their fears and failures and ultimately, their successes.
Pretty soon, you feel like you know this person sending you these emails.
That’s the power of being personal.
And while I can’t teach you to have a personality, I can give you some pointers:
Start your emails with brief, relevant anecdotes
Use a conversational tone in your writing
Talk about real failures and successes that you and your customers have experienced
Always be relevant to your audience
Being personal is not hard, but it does require you to get out of “marketer mode” from time to time.
7. Authenticate your Email Sender: SPF, DKIM, DMARC
If your emails are not reaching the inbox of your subscribers then your email will never be opened. If you want to improve your email open rates, prove to the servers – particularly big corporation servers – and subscribers that it is really you who is sending out the emails.
Here’s what you’ll need to do to get your email authentication going.
Technically, a “bounce” is a visitor that looks at only one page, or a visitor that spends an embarrassingly short time on the page. Keep reading to find out how to reduce bounce rates.
A bounce is any visit for which the visitor only looks at one page and does not interact with it. This sounds truly unfair as someone may spend minutes on your blog post or landing page, and still be counted as a bounce.
A visitor bounces when they don’t find anything close to what they were looking for when they visit your site. Either you’re attracting the wrong visitors or you don’t know why they are visiting.
Bounce is the most extreme form of conversion problem. High bounce rates are an indication that you are throwing good marketing dollars down the tubes. Whatever you’re spending to get traffic to your site is being wasted.
How to Reduce Bounce Rates or the heartbreak of “bounce”
Boing!
That’s the sound of someone finding your site, but not finding what they wanted ON your site.
Boing!
That’s the sound of website content that doesn’t match your marketing.
Boing!
That’s the sound of a website that talks about the company instead of the visitors’ problems.
Bounces Aren’t Helpful to Businesses
What are some strategies to reduce bounce rate?
This is a common question, and requires an understanding of the definitions of bounce rate.
The bounce rate is a bit slippery and requires some examination. The intention of measuring the bounce rate is to figure out how many of your visitors are leaving almost immediately after arriving at your site. This metric provides for a lot of error in interpretation.
“A high bounce rate means your site is crappy.”
This is rarely the case. A more accurate explanation is that your site doesn’t look the way your visitors expect it to look. Understanding what your visitors expect is the way to reduce bounce rates.
Instead, there are usually some more valid reasons for your high bounce rate. Here are the things digital marketing and conversion experts examine when confronted with uncomfortably high bounce rates.
1. You’re measuring it wrong
How you measure your bounce rate can give you very different insights. For example, blogs often have high bounce rates. Does this mean that visitors don’t like the blog?
Many analytics packages measure a bounce as a visit, or session, that includes only one page on your site. Visitors who take the time to read an entire article would be considered a “bounce” if they then left, even though they are clearly engaged.
We set a timer for our blog traffic, so that any visitor who sticks around for 15 seconds or more is not considered a bounce. You can set a timer to the amount of time you consider appropriate.
2. How to Reduce Bounce Rates: Diagnose Technical Difficulties
We are fond of saying that you don’t have one website, you have ten or twenty or thirty. Each device, each browser, each screen-size delivers a different experience to the visitor. If your website is broken on one of the devices popular with your visitors, you will see a bump in overall bounce rate.
If your pages load slowly, especially on mobile devices, you can expect a higher bounce rate.
Broken internal links and 404 pages are also cause for bounce.
If your page breaks out in a chorus of Also Sprach Zarathustra when the page loads, you may enjoy a higher bounce rate.
How to diagnose device-related technical problems
Your analytics package will track the kind of device your visitors are coming on.
Is there a problem with this site when viewed with the Safari (in app) browser?
The Google Analytics report Audience > Technology > Browser & OS shows that there may be a technical issue with Safari visitors coming from within an app. This may also reflect visitors coming from mobile ads, and they may simply be lower quality. See below.
With Google Analytics Audience > Mobile > Devices report, we see mobile devices specifically. The Apple iPhone has an above-average bounce rate, and we should probably do some testing there, especially since it’s the bulk of our mobile traffic.
With an above average bounce rate, visitors on an Apple iPhone may be seeing a technical problem.
3. Good Traffic Quality Helps Reduce Bounce Rates
If you are getting the wrong visitors, you will have a high bounce rate.
Remember StumbleUpon? Getting your site featured on the internet discovery site often meant a flood of new visitors to your site… and a crash in your conversion rate. Stumble traffic was not qualified, they were just curious.
Your bounce rate is a great measure of the quality of your traffic. Low quality traffic bounces because:
The search engine showed them the wrong link or a broken link. Do you know how many visitors used to come to our site looking for a “conversion rate” for Russian Rubles to Malaysian Ringletts?!
User intent. The visitors aren’t ready to buy. They were in a different part of the purchase process. Visitors coming from Social Media ads have notoriously low conversion rates. They weren’t looking, they were just surfing your product pages.
We take a closer look at the source of traffic to diagnose a traffic quality problem using Google Analytics Acquisition > All Traffic > Channels report.
Direct is one of our biggest traffic sources and brings in one of the two highest bounce rates.
Here we can see that traffic coming from social media and those visitors coming “Direct-ly” have a high bounce rate.
If you are driving a lot of visitors to your home page, you may want to consider presenting them with links to more relevant content. As Tim Ash says, “The job of the home page is to get people off of the home page.” He didn’t mean by bouncing.
With regard to social media, we may have a problem with broken promises.
4. Broken Promises Lead to Conversion Problems
Do your entry pages consider the source of visits?
If your traffic is clicking on an ad that promises 20% off on a specific propane grill, and they’re directed to your home page, you’ve broken a promise.
You might think that they will search your site for the deal. You might even think they’ll search your home page for the deal. You’re wrong. Many will jump.
Every ad, every email invitation, every referral link is a promise you make to your visitor. If they don’t come to a page that lives up to the promise, they are likely to bounce.
Does the headline on the page match the offer in the ad?
Does the product in the email appear after the click?
Are your calls to action in alignment with the landing page?
Are the colors and design consistent across media?
This Ad takes the visitors to a page that is almost designed to disappoint.
Looking at your ads on a page-by-page basis is necessary to diagnose and correct this kind of bounce rate problem.
5. Vague Value Propositions don’t help reduce bounce rates
Ultimately, if you’re not communicating your value proposition to your visitors clearly, you are going to enjoy a monstrous bounce rate.
21 Quick and Easy CRO Copywriting Hacks
Keep these proven copywriting hacks in mind to make your copy convert.
43 Pages with Examples
Assumptive Phrasing
"We" vs. "You"
Pattern Interrupts
The Power of Three
"*" indicates required fields
Your value proposition typically does not address your company or your products. It should be targeted at your visitor, why they are there, and why they should stick around.
Each page has its own value proposition. Your business may have a powerful value proposition, but each page should stand on its own.
A contact page should talk about what will happen after you complete the form. Who will contact you? How long will it take? Will they try to sell you something?
A landing page should clearly state that you are in the right place and provide reasons for you to stay and read on.
A home page should help you find your way into the site. Most home pages are treated like highway billboards. No wonder people just drive on by.
6. Finally, Is Your Stubborn Bounce Rate Mocking You? [AUDIO]
Most of us, at one time or another, have not only been frustrated with our bounce rate, but completely puzzled about what to do about it.
In this podcast, I will guide you toward a more accurate bounce rate. I will walk you through the steps to use The Timer Listener to time the visit duration of each of your visitors and get those percentages down where they belong.
Subscribe to Podcast
Ultimately, we don’t want to reduce our bounce rate. We want to improve our conversion rate by bringing the right traffic, to the right page, with the right message, and avoid technical issues that get in the way.
How to Compare Data When You Move from Google Analytics to GA4
Web AnalyticsOnce you move from Google Analytics to Ga4, you can be assured that your data will not match exactly. In fact it may be off by a percentage.
What is more important is that relative changes from day to day, week to week, and month to month are of the same magnitude in both systems.
Is the GA4 data correlated with the UA data?
For example, if we graphed the Universal Analytics (UA) metric Users against the GA4 metric Active Users, it might look like this:
It is clear that UA consistently reports more Users than GA4 active users on a day-by-day basis.
The blue bars are the users reported by UA and the green bars are the active users reported by GA4 on a daily basis. It’s clear that UA is reporting more users than GA4 is reporting acitve users.
This is to be expected, because active users are calculated differently by GA4 than is users.
What is more important is that they move similarly to each other day after day. In other words, if GA4 is going to report fewer active users, the magnitude of the difference between it and UA should be consistent, day after day.
For most days this appears to be true. But some days, UA reported many more users than GA4 reported active users.
Does this mean that we can’t trust one or the other? There is a way to find out.
Scatter Plots, Not Bar Graphs
The bar graph is a crude tool for comparing two data sets. In fact, any time-series graph is going to disappoint.
What we need is a Scatterplot.
The Scatterplot graph in Google Sheets
A scatterplot ignores the order of the date and instead compares the data on each day. On a day that UA reported 200 users, how many active users did GA4 report? We plot that point.
When we do it for each day in our data set, we might see something like this:
Scatterplot graph of Users from UA vs. Active Users from GA4
What you might notice is that this data lies in a straight line, for the most part. This is a good sign. It means that the GA4 data changes relative to the UA data for each of the days mapped.
This doesn’t mean that it’s accurate, though. Here’s a scatterplot of the same data, but I’ve artificially doubled the daily UA data.
A scatterplot in which the value of one dataset has been doubled. It looks pretty much the same.
This data looks good, but it’s not. How would we know?
Spreadsheets and your high school math teacher give us a simple way to evaluate the data like a boss.
Add a Trendline
First, Google Sheets will calculate a trend line for us. When at science events, we call this a linear regression. This is the straight line that best “fits” the points. If the points look like a line, then the trend line will be a close approximation of the data. In Google Sheets you’ll find this in the Customize tab under Series >.
These features exist in Excel as well.
Check the Trendline box in the Chart Editor of Google Sheets
When we add a trend line to our data, we see this:
The trend line matches the data very closely. But how closely?
That draws a pretty line right along with our data. How closely do the two data sets match? That’s what R2 tells us.
Reading the R2 Value
If you’re curious about how this is calculated, here’s a helpful video.
Google Sheets will calculate R2, but this is not enough. We want the equation of the trend line so that we know how closely related the two data sets are.
In Google Sheets, set the Label field to “Use Equation” and check the box labeled “Show R squared”.
There are some mathy looking bits in our legend now.
The R squared value tells us how closely the data “fits” our trend line and the equation describes the trend line in detail.
The R2 number tells us how well the trend line describes our data. A perfect fit would give us an R2 value of one. The closer to one it is, the more likely our two data sets are describing the same thing.
The equation is the one you learned in high school. It’s just the equation of a line.
The Equation of a Trend Line
This is one of those equations that you swore you would never use in math class. Today, it’s going to give you X-ray vision into your data.
x is the GA4 Active Users
y is the UA Users
The choice of x or y axis is arbitrary for a scatterplot.
m is the “slope” of the line. It’s the “rise over the run”. If we expect our two datasets to be alike, we expect a slope very close to one.
B is the “y intercept”. It is where our line crosse he vertical axis, also called the “y axis” when x is zero.
We’re hoping that our GA4 data is as much like our UA data as possible. If the two were reporting the exact same number each day:
I compared two identical data sets to show this.
Two identical data sets. The R squared value is 1. The slope is 1.
The y-intercept is 0.
So, what if our data isn’t perfect?
If R2 is significantly less than one, the two data sets are not well-correlated to each other. In other words, they are not describing the same thing. If it’s 0.9 or above, we feel pretty good about the comparison. If its below 0.8, we should be worried.
Even if R2 is close to one, the slope (right before “x”) might be significantly less than one. In this case, we would find that that one dataset is adding or subtracting a percentage of the true value. It could be doubling the count of users, or not reporting users on some percentage of the pages of your website.
If the R2 value his close to one and the slope is close to one, we may find the y-intercept to be higher than zero. This means that some consistent value is being added to one or the other dataset. One is counting something that the other is not.
Here are some common scenarios we see in comparing UA and GA4 data, and how the equation would be expected to change.
You’re comparing the wrong data.
Let’s start off by looking at a bad correlation. Here the R2 value and slope are near 0. The y-intercept is very high.
Something is just not right here. Maybe you’re not pulling the data right.
Both the R2 value and slope are near 0. The y-intercept is way above 0. This is data that doesn’t correlate.
Bot traffic is not being filtered in one dataset.
In this example, I’ve artificially added 50 users per day to one of the datasets. This is what it would look like if GA4 was filtering out a consistent traffic source, like bot traffic, but UA was not.
The entire trend line will is lifted by 50 users. Because it’s consistent, the slope and R2 values are not affected. But the y-intercept will rise precariously.
Adding 50 users to one dataset increases the y-intercept, even though the slope and R2 values are near one.
You’re double counting.
It’s remarkably easy to double-count by adding the Google Analytics tag twice. In this case, the slope will be close to 0.5 (or 2.0 if you flip the x and y axis in your scatterplot).
It’s not unusual for us to find a website that is adding pageviews using an on-page tag and a tag manager tag. This will double-count pageviews.
When you double the users reported in one dataset, the slope will approach 0.5 or 2.0.
You are “breaking” sessions.
If you are “breaking sessions” in either dataset, you’ll see inflation of sessions. This will be reflected in the slope. It will be significatnly above or below one.
For example, if you use a utm_ query parameter on a call-to-action button on your site, UA will start a new session, as if the user was just arriving on the site. GA4 doesn’t do this.
If your visitors are going to a third-party site and returning, you can get broken sessions. If you have cross-domain tracking setup in UA but not in GA4, you’ll see something like this for the segement of visitors that visit the other site.
The analytics tag is missing on some pages.
With this example, I’ve added 50% to the dataset on the Y axis. This simulates the scenario in which 33% of the pages on the X-axis dataset don’t have tags.
Note that the R2 value doesn’t change. However, the slope of the line is well below 1. In fact, it’s about 2/3 of a perfect slope.
When adding 50% to one dataset, you find the slope changing even though the R-squared value is the same.
Revenue, Transacations and Segments
This approach can be used to check most of your metrics and segments.
Not only can you evaluate the data you are collecting, you can evaluate your ability to pull data in GA4 that represents the thinking of the UA developers. UA data is pre-processed differently in UA than it is in GA4.
This is a great way to be sure you’re pulling similar data segments.
Compare Google Analytics to your sales data.
If you want to be sure Google Analytics is collecting ecommerce data, you can compare transactions from GA to transactions from your backend, such as Shopify, BigCommerce, Magento, etc. This approach is great for that.
This is one of the first things we do with or new Conversion Catalyst clients.
The graphs look the same. Don’t be fooled.
Be careful when you move from Google Analytics to GA4.
In all of these examples, the scatterplots look pretty much the same visually. However, our high school math teacher has equipped us with the equation we need to diagnose our data.
Thanks, high school math teacher!
CRO and SEO Working Together with Jason Fisher
Conversion Marketing StrategyWhen CRO and SEO work together, there is a cycle of increasing advantages, according to technical search engine marketer Jason Fisher. Find out how Jason combines SEO and CRO for a one-two punch that delivers results.
Jason Fisher and Brian Massey
Subscribe to the Podcast
iTunes | Spotify | RSSAll Episodes
What if you were Google Search’s therapist? You would be tasked with understanding the mind of Google’s search algorithm, simultaneously seeking to understand the Google mind, and trying to help it make better decisions about the world it manipulates.
This, as it turns out, is the job of the search engine optimizer, or SEO. Maybe we should call them SETs: Search Engine Therapists. For they must not only understand the search algorithm, but they must also be able to help it make better decisions.
And they must do this with the active resistance of Google. Like humans, Google secretly does not want to be helped.
Our intrepid SETs will never have a complete understanding of the Google search mind. So, we are stuck with a simple mantra: “Take the best search traffic we can get and let the website sort it out.”
I’m the guy that focuses on “sorting out” the traffic. Making a website better at finding the buyers in your traffic is called Conversion Optimization, or CRO.
Here lies the delicate balance between making the search mind better at its job (SEO) and making the most of what comes your way (CRO).
Google is a cantankerous patient who makes the therapist pay for the privilege of helping it. We need to use all our weapons to maximize this traffic source.
This is why I invited Jason Fisher onto Intended Consequences. He is a search therapist who gets that the “sorting out” part of the equation is important.
* * *
Jsason Fisher has been doing search marketing for a long time. He cut his teeth on search engine marketing right at the beginning of the century, when search marketing was the like the Wild West. Some years later, he was working for a leading link development company, where he learned the power of back links.
He is filling the technical SEO skills gap, primarily for agencies. And he’s the kind of guy that throws around phrases like “competitor link graphs”.
I was curious how a guy like this sees conversion optimization in relation to his work. Here’s what I learned.
CRO vs. SEO: Advice to new online businesses.
We do a lot of consultations with businesses that aren’t ready for our conversion optimization services. My advice to them is this:
Organic search optimization takes 12-24 months to implement. It takes time to research, to create valuable assets, and to build relevant backlinks. But this cannot be easily taken from the business.
Paid search can begin working immediately, but is subject to the bidding of your competition and the whims of the search engine ad algorithms.
Of course, you don’t have to put conversion optimization on hold during this time, especially if your paid search campaigns aren’t yet profitable.
What are the components of SEO?
For established websites, Jason discusses four main components of an organic search program:
Portion of a hamburger with label “On-page/Off-page Factors”
These on-page and off-page success factors are the things the search engines look at to determine your authority for certain queries.
Host performance is an on-page factor. If you have a good tech stack, you should be good.
Over-indexing and Under-indexing
An often overlooked aspect of your search strategy is that the wrong pages are being indexed.
Over-indexing and under-indexing are important CRO and SEO factors.
Do you have a lot of poor-quality pages indexed on the search engines? This can be a problem.
Do you have quality pages that aren’t indexed? Of course, this should be addressed.
It may seem counter-intuitive that you should remove some content from the search index. Jason says that there is a limited “crawl allotment”. You may be wasting yours on poor-quality pages by marking them as “no index”.
Where does link development fit in today’s digital marketing practice?
Link building starts with creating valuable assets, things that others want to link to. In this sense, Jason sees link development like public relations: you’re trying to get people with authority to link to your work.
It takes a lot of work. Of those solicitation emails that us publishers get daily, only 1% results in a backlink.
Evergreen Content
Content is the meat of the sandwich.
Evergreen content are those assets that can stand the test of time should be a focus of your efforts.
Update them and relaunch. Always be looking for ways to increase the quality. For example, transcribe videos on that page to text.
Content Blueprint
Start with the top keyword opportunities in your space. For those phrases that you don’t have content for, you can retarget existing content to them, or create new content that addresses them.
You are closing the gap on the most important search phrases for your business.
Your goal is to demonstrate to “the machines” that you are the authority in your space. People are going somewhere to get answers to questions relevant to your business. Why not your website?
Some of these topics may seem counter to your intuition. Why put DIY content on your site when you want to do the work for them? Because it tells the machines that you are the authority for those search phrases.
Can we outsource this content?
It depends on how complex your product or solution is. You may need a subject matter expert to truly develop quality content.
Content that gets visitors to stick around.
Content that gets others to link to it.
How can a website convert more visitors to prospects and customers?
Your website needs to
People overthink product and service pages. Jason likes to focus on those things that differentiate your offering.
Yes, testimonials and customer logos can increase your conversion rate. Yet, selling your brand and it’s values make a big difference.
This is harder than adding logos to your page.
It is especially difficult to do your own writing. We are too close to our businesses to present them in a way that the visitors need. Let an external resource start the page and then review it to ensure it reflects your advantages and brand voice.
All of this is the foundation for a great marketing program.
Search optimization provides the information that can be used across your marketing efforts.
You will learn a great deal about your audience and your industry when you look through the eyes of the search landscape.
What do your prospects want? What are your competitors doing? What is missing from your messaging platform?
The content calendar is a marketing roadmap.
Understanding not only what you need to say, but when you need to release messages are key to success in search marketing and marketing in general.
When testing your website, you need to know when to run high-risk tests versus low-risk tests.
For example, during peak holiday traffic, you can learn much from low-risk testing during this high-traffic timeframe.
Write for the machines or write for the visitor?
Jason says that you always write for the visitor. Your selection of topics may be driven by the search engines’ understanding of your industry, but the content is for those entering the search phrases.
Links
Jason Fisher is an organic search marketing veteran with over 15 years of experience. Jason has helped numerous Digital Marketing Agencies all over the country increase their clients organic search channel sales & revenue by building & executing sound SEO Strategies.
Jason Fisher’s Website
Why Web Design for Conversion Needs Science
Conversion-Centered DesignWeb design is about communicating with people who have problems that your offering solves. Science is about making sure you’re not designing just for yourself. If you do web design for conversion, science will guarantee that your new design will perform better. Here’s how.
This is Alice. She’s a web designer. Alice is doing a website redesign. The design she completes is based on research and her extensive experience. She is confident that this new design will make more visitors choose her company.
Alice the Web Designer
Alice works with Bob, the web developer and Cindy, the marketing manager to finalize her design, copy and images.
Alice the web designer, Bob the web developer and Cindy the marketing manager.
Their boss, Doug, has faith in the team and thus the design. He loops in Emily from Sales. She thinks it is a big improvement.
Yet, we haven’t asked any customers yet. So they invited some prospective customers in to weigh in on the new design.
Most of them thought the new design was a definite improvement.
The focus group was mostly in favor of the new design.
Frank, who worked in reception was concerned. He loved to shop online, and knew that if a product had a 5-star review but had only 12 reviews, he wouldn’t believe that rating.
So, why was 12 opinions enough to believe that this design is actually an improvement? Five of them were company employees, after all!
How 12 people can get it wrong.
Fortunately, Georgia, the company scientist was also concerned. She knew that the internal group was going to love the new design because they had designed it. And he knew that members of focus groups want to like the new design, because they are eager to please.
They’re human.
Instead of replacing the old design with the new design, Georgia setup an experiment. Half of the visitors to the website saw the old design and half saw the new design. Which one would generate the most revenue? What she found was that there was no difference. As far as the company’s prospective customers were concerned, there was no difference.
How did a talented bunch of people, with the help of real customers come to such an erroneous conclusion?
Let’s assume we have 100 people visiting a website. Half of them are more likely to buy with the old design and half of more likely to buy with the new design. You can see that it’s pretty easy to collect a sample of opinions that don’t tell you what is really going on. Pick the wrong 12 people, and you can either bet on a poor design, or throw out a pretty good design.
Most design teams are biased in favor of a design because they created it with no other opinions involved.
Bigger sample sizes mean fewer design mistakes.
We can never ask EVERYONE if they prefer to see the old design or the new one. We have to ask a subset of the population of visitors. Scientists use the variable “n” for the size of the sample we are going to “ask”.
Here’s the problem from a Design Scientist’s point of view.
One designer? n=1
One designer, one developer, one marketing persion? n=3
Add in an executive and a focus group? n=12
At n=12, we are till making mistakes in our design decisions. How big does n need to be?
In our example, a sample size of n=40 gives us more confidence that we are seeing reality.
It’s harder to make bad decisions with larger sample sizes.
Where do we find these bigger sample sizes?
A design scientist has two tasks:
There are three broad ways of getting more n’s to look at your design options:
Usability Testing
A usability test is essentially a giant focus group. Thanks to services like UsabilityHub, you can bring 25, 50, or more people to look at your new designs and tell you which communicates better.
Cons: These panels of people are NOT necessarily customers of your product, so their input is less reliable than trial and error or an AB test.
Trial and Error using analytics
To get the “input” of your actual prospects, it makes sense to launch something. Then you use analytics to see if the change made things worse or better. However, you must be willing to roll your new design back if you find that conversions drop with the new design.
We recommend using an AB testing tool to make changes to the page and then use analytics to determine if the change was an improvement. If it was, make the change permanent. If it wasn’t, try something else.
Cons: If there is a shift in traffic, pricing, promotions, competitors, or anything else, your results can be skewed. For example, if your competition launches a sales at the same time that you launch a new design, it can look like the new design is decreasing your sales.
AB Testing
The AB test is designed to overcome the limitations of the other two approaches. It takes it’s sample from your actual web visitors and it controls for changes in the marketplace. For an explanation of how this works see our intro to AB Testing.
Cons: AB testing is limited by the amount of traffic you are getting and requires some developer support.
Guaranteeing your design will outperform the old design.
If you are able to improve the number of brains involved in your web design for conversion, and can bring the brains of actual prospects, there is no reason you should make small sample size mistakes with your website redesigns.
You can guarantee your new design will improve conversions. If your samples say they’re not better, you don’t use them and try something else.
What are the Steps of Conversion Optimization
Conversion OptimizationBrian Massey, the Conversion Scientist, shares the steps of conversion optimization. He is the founder of Conversion Sciences, and author of the book “Your Customer Creation Equation”.
If you are new to conversion optimization or if you need to take your website conversion to the next level, Brian’s book is a fantastic foundation. It will help you understand the way an optimizer looks at the world and looks at a website.
What is the Goal of Conversion Optimization or CRO?
Let’s start off by talking about what conversion optimization means to me. I don’t see my job as just improving conversion rates, but getting the most value out of every visitor to your website.
In a lot of situations that might mean a conversion to a sale. But even on an e-commerce site, we might want to connect with visitors who aren’t yet ready to buy, so we try to get them to join an email list. We can get value from them by asking them for their name and email address in exchange for fantastic information, a discount, or something else of value.
In the mobile world, a conversion may look like a phone call. Click-to-call is a powerful way for prospects to take action when away from their phone.
There are a number of things we can do to get value from our visitors, and for us, everything is on the table.
The reason I want to lead the life of a conversion optimizer is because we do wonderful things for online businesses. This is probably why we’re so well-liked.
Benefits of conversion optimization.
Benefits of Being or Having a Conversion Optimizer
First of all, we do increase the revenue from the traffic that you’re getting. The net result of that is that it decreases your acquisition costs, your advertising spend.
If more of those clicks that you’re paying for turn into customers, then you get a positive return on your ad spend.
Having a site with high conversion rates often means people are staying on your site longer. They are buying more often. They are typically visiting more pages. Google understands this and rewards you with higher rankings. If people are staying on your site and not “pogo sticking” back to the search results page, well, your organic ranking will likely go up.
We, of course, increase your growth because we don’t just test what goes on your pages. We can test pricing, we can test bundling, we can test new features — we can test the things that are core to your business.
And we provide the data that makes you smarter at identifying where you should be investing your advertising and marketing dollars.
We’ll know which channels are converting best, which things you’re doing well and which you are not. And you can adjust your marketing spend accordingly.
So, conversion optimizers really are wonderful for a business.
Do you know which path your users are taking?
Logically, we might think that a path through our website should look a certain way, but in truth, the visitors want something different.
It’s our job to understand what that desire path is. For instance, in a park like this, people may be avoiding the paved path because it’s concrete which is harder on your knees when you’re running than dirt. It may not just be because it’s the shortest distance.
We want to understand the visitors’ motivations. That is what I spend my days doing. My whole job is to make sure that I’m not using mental shortcuts to make decisions.
Cognitive bias codex.
The 3 lbs. of gray matter between our ears is just packed full of biases, shortcuts, and stereotypes. These biases, stereotypes and shortcuts, cause us to think we’re doing the right things when we’re making decisions about design, or about our products, or about our pricing.
But in truth, we’re doing it wrong for our users. We take shortcuts. We’re not really connecting with what our users want. This whole method, the steps to conversion optimization, is designed to keep us from relying on our biases to make decisions.
The Importance of the Optimizer
When you’re optimizing, you play a really important role in the design process and in your company. You are the one who is double checking the assumptions that are being made. You are the one making sure that those assumptions are what our visitors and our customers want.
The optimizer plays a key role in the design process and in their company.
The Benefits of Conversion Optimization
An optimizer has many benefits. They save time. There’s nothing that wastes more time than launching a campaign, spending your marketing budget on that campaign and then not having it work. So, by doing a little extra work on the front end, collecting some data, you can make sure that your campaigns are going to be more successful so you don’t have to start over and relaunch them.
The Value of Data (and its many uses)
Data is a great way to deal with what we call helicopter executives, executives who feel that maybe the team isn’t making the right decisions. They feel that they have to come in and review your creative and your campaigns, making changes to what you’re doing. Of course, their assumptions are based on the same biases that anyone’s are.
If you are able to say, “Well, we have some data that says this is the best thing,” then they’re more likely to think, “OK, this makes sense. Go ahead.”
You’ve just removed one cook from the kitchen.
How a conversion optimizer should handle agencies and teams.
Oftentimes you’ll get creative from your agency and think, “Is this really effective creative?” Your agency may present you with options and ask you to choose. As an marketer, your answer should always be “I don’t know, go collect some data to find out which one of these ideas is most likely to be the best choice”. This should be the job of your agency’s experimenter. This is a powerful way to manage teams effectively.
Steps to Conversion Optimization: Gathering Good Competitor Ideas
We like to take ideas from our competitors and from other websites that we like, but we often steal bad ideas. Just because our competitors are using them doesn’t mean they’re working.
An optimizer wants to take ideas and test them before stealing them. At Conversion Sciences, we say, “Steal like a scientist.”
Digital Marketing Careers Require Experimentation
If I haven’t made the point abundantly clear, people with the skills of an optimizer are very valuable. And right now these skills are hard to find and expensive. In a few years, these skills are going to be absolutely required. So, if you don’t have these skills, you’re not going to be able to work in premiere digital marketing roles, in digital product management, or run a business that requires the web to succeed.
Experimenters Can Take More Chances
As a conversion optimizer, you can take more chances because you know how to create experiments that allow you to be more creative.
Experimenters take these really creative ideas that would otherwise sound risky, find a way to collect some data, and then understand whether or not that idea is actually going to improve things. You also avoid implementing a bad idea. We call this a “design insurance”.
You don’t have to always play it safe with your campaigns. You can come up with crazy ideas and experiment before you actually launch and take all of the risk.
Being able to take more risks, a CRO expert can get more leads, more sales, and lower acquisition costs.
And, of course, you get more leads, you get more sales, you lower your acquisition costs, you grow your business.
That’s what most people want from their conversion optimizer. But conversion optimizers are so much more valuable.
My day deals almost exclusively with ideas. Ideas for how to improve a website, ideas for how to improve a customer’s journey, ideas for what kind of content we should be putting on the page, ideas for how we should discount, ideas for how we should lay out a page. Ideas for what we should be doing and advertising. For almost anything that’s going to be seen or experienced by the user, there are ideas for improving it.
How to Find the Right Ideas: good reasons to kill ideas
I’m going to walk you through the process of figuring out which ideas are the right ideas.
When we first start with a client, we go through their website and perform an analysis. This includes an analysis of their existing data. We come up with a very long list of what looks like really good ideas for improving the conversion rate.
Our job is to kill some of those ideas and get them off the list so that we can move on to the ones that are good ideas. In fact, the scientific method that I use on a daily basis is designed around this.
The job of an experimenter is to come up with ideas and then find out why that idea is wrong. When you test a hypothesis, you are actually testing against the null hypothesis trying to prove that idea won’t improve things. If you can’t, despite trying everything, then you’ve got a winner.
So, what are the good reasons to kill ideas? We evaluate ideas based on these criteria. Is there a reason that we should keep this idea on the list?
Reasons to Kill an Idea
1. That’s a lot of work
Some ideas require too much work to test and implement. We might say the website needs to be redesigned. That’s very risky because it changes everything. And so we’ll often just pull that idea out right away.
Ironically, this is the way most website redesign is done. 90% of the market is still redesigning websites this way.
They start by hiring a creative agency or bringing in a creative team. That team does a little research at the beginning of the process, and then they make all sorts of design changes based on that research. Then, they push it all out and hope that they made the right choices.
That’s very risky, and so full-scale redesigns don’t stay on our list very long.
2. It’s too small of an idea
Some ideas just aren’t that impactful. For instance, if we had an idea to change something in the footer of a page, we can tell from our heatmap reports that few visitors are seeing the footer area. We would say that’s too small of an idea. It doesn’t have enough of an impact and we’ll drop it from the list.
Likewise, changing the color of a button or changing the font of our headings are low-impact changes. We tend to just get rid of these ideas.
3. No one is seeing it
There are pages on your site that are important to the customer’s journey, but not a lot of visitors are visiting it.
For example, sometimes FAQ pages can be really important to our visitor’s journey. If we had a hypothesis that said we’re going to change the order of FAQ questions, but we looked in analytics and saw that few visitors were actually visiting the FAQ page, we would say it’s probably not a good thing to test.
On the other hand, few people are seeing the checkout process on an ecommerce website, but those visitors are in the process of buying. In this case, we want to keep checkout ideas at the top of our list.
4. I don’t have any data on it
For each idea on my list, I have to ask myself, “Can I find some data on this idea.” This is the question we ask ourselves over and over and over. If I can’t find data on an idea, or I can’t generate data on that idea, then it’s not a testable idea.
A good example might be things on a website that encourage people to visit a physical store. There are technologies to track this cross-channel behavior, but it’s very expensive technology. Even if we have really good ideas about how to drive more people to the brick and mortar store, we really don’t have a way of collecting success data related to that idea. So, that would be something that we would eliminate because we don’t have the data.
Steps of Conversion Optimization: Gather Existing Data
Let’s talk about sources of data. Once we’ve gone through our list, we’ve got things on it which we think are good ideas. We think they are easy to implement or can be implemented in a reasonable amount of time. We think people are visiting those pages, and we think we can find some data on them.
One of the first places I like to look when I’m building a landing page are the client’s paid search ads.
Steps of Conversion Optimization: Gather Existing Data.
Using this made-up example for the U.S. Mint — that’s the part of the US government that prints money — let’s pretend that they’re offering 50% off dollar bills.
Now, you might think this is a crazy offer, better than anything you have. But the truth is that we all have an amazing offer: a great product or service that’s priced right. It saves time. It saves money. It solves a problem. Yet, you still have trouble converting people. Well, don’t be too discouraged, because the U.S. Mint would have trouble giving a dollar bill away for 50 cents.
Go to your paid search team or your advertising team and ask them for a spreadsheet with the last six months or a year’s history of ads.
Go through the data and look for those ads that had a lot of impressions or more importantly, had a lot of impressions AND conversions.
If we look at the third one in our example above, we see it has an interaction rate of 2.8%. That seems like the highest rate, but it was only 37 interactions. This sample size is a little bit too small for us to have confidence in. I’m more interested in those that have 612 or 943 conversions.
It seems that “50% off dollars for a limited time” has a better conversion rate than “Dollar Bills: Buy one, get one.” When I write copy for my landing page, I’m going to favor language that includes “50% off”.
I would not be as excited about “Discounted Dollar Bills” because it had a 0.3% conversion rate and a high enough number of interactions that we can believe that this data is probably accurate.
You see how I can use our ads to understand which words, which headlines, I should be using in my landing pages and in my copy. It’s on my list.
Social media Ad Performance and Conversion Data
We can do this with social media as well. For instance, if we want to put a video on the landing page or video on our homepage, video ads can help us understand what people are interested in.
Use a tool like Ahrefs or SEMrush to look at our competitors’ ads. Find out what words they’re using and if they have a lot of keywords that they’re using.
Previous Email Campaigns Data
Email campaigns are another great source of data. I looked at the subject lines for emails that Conversion Sciences sends. We send emails for new blog posts all the time. I wanted to know which subjects — which titles of our blog posts — were getting the most clicks. I took six months of data and I ranked it based on their click through rate.
Previous email campaigns data: email subject performance.
Looking at the top ten, we see “writing”, “copywriting”, “persuasion”, “value proposition”, “persuade people”, and “business taglines”. My audience is interested in the words that influence conversions.
I was a little surprised, but we were able to use this data to produce a free copywriting report on how to write copy for conversion writing, and this converts very well for us.
Download these 21 quick and easy CRO copywriting hacks.
In this case, the data really did point us in the right direction. The data created a hypothesis, an idea, and then gave us the data that said you should launch this. Then we used the conversion data, the number of leads that we’re generating on our website, as the final proof that this was a good idea.
Steps of Conversion Optimization: User Testing
Another step in conversion optimization is user testing. Everybody thinks that a conversion optimizer spends most of their time split testing. This is the best data we can generate, the best tools that we can use. But in truth, I want to gather data faster and doesn’t require me to use precious visitor traffic.
We only want the most important and best ideas to go to AB testing while using user testing to figure out which of my creative ideas is best.
Our user testing includes things like a 5-second study. A 5-second study works great when I have three or four different headlines and three or four different images that I want to consider for a landing page.
We’ll use a service like UsabilityHub or Helio and we’ll ask for 25 people to come and look at each of our mockups. The 5-second test works like this: test subjects get to see the mockup for five seconds and then it disappears.
But five seconds in the human brain is quite a long time. After the five seconds is up, we’ll ask questions like,
We can score these twenty five people in each of these areas. The image and headline combination that scored the best tells us that it’s probably the best idea.
We now have some data from real world people that is telling us which idea to take to an AB test. There might be a couple of these that score well. So, we want to take the two best ideas to an AB test, but it also means we don’t have to test the others and waste traffic on those.
There are a number of tests that you can use for user testing. Usability hub or Helio, offer a question test where the visitor gets to look at the page as long as they want and answer questions.
A first click test measures how quickly someone can find where they’re supposed to click based on the prompt that you give them. How many of them get it right in test layout or how clear the call to action is on your page.
User testing tools like UserZoom or UserTesting.com allow us to set up a scenario and ask the visitor, for instance, to go through and purchase on a website. We watch them as they try to complete the task. We see where they get confused, where they get tripped up. They’re talking out loud as they’re going through it.
You’re going to see issues in these user tests that you wouldn’t catch otherwise. It can be very enlightening. We can really learn quite a bit from that user testing videos.
More Data Sources: User Intelligence Tools & Reports
Another step of conversion optimization is to look through user intelligence reports. User intelligence is different from user testing.
User testing uses strangers and pretenders. These are people who aren’t actually trying to solve a problem, but we’re using them as a focus group to play with our creative and see how effective it is at communicating with human beings.
User intelligence tools are actually watching your visitors as they interact with your website.
Analytics has the most obvious user intelligence data. Google Analytics is a great behavioral database. It’s all the people who are coming to your website to try to solve a problem. It shows you where they landed, what channel brought them, what pages they visited, how long they were there, where they left, if they bought, how many of them bought, what their computer setup was, what browser they’re on — all of this information is in Google Analytics.
It’s a great database for asking questions. I probably spend at least 10 to 20 percent of every day in analytics, and if I’m working on an analysis, I’ll spend the entire day in analytics, it’s such a rich source of data.
The other thing we use is what are called heatmap reports. They tell us how far the visitors are scrolling when they visit a page, where their mouse is moving on a page, and where they’re clicking. These are great tools for answering specific questions about a page.
You don’t have to be a Ph.D. in science to understand them. If you can read a weather radar map, you can read a heat map.
Here is an example.
Heatmaps of a website page for golf resort in Hawaii.
This is a resort in Hawaii, a golf resort. They assumed that since it’s a golf resort, people who are considering booking a room are going to be interested in golf.
On this page, which lists all their specials, they listed the golf specials at the top. When we went in and looked at where people were clicking on this page, we found out that “Free Breakfast” was most clicked item, even though free breakfast is down near the bottom of the page.
People don’t behave the way you think. What is the cost of breakfast? At this resort, it might be 40 or 50 bucks. If you’re going to save a couple of hundred dollars on golf, it would seem to be a better value, right? Not according to the visitors.
These are the sorts of insights that conversion optimizers love to find.
I also spend time watching session recordings. With session recordings you get to watch visitors as they’re working through your site. You see where their mouse is moving and what they’re clicking on. It takes a while, but you find things that you wouldn’t otherwise discover.
Session recording of golf rates page.
If you watch a bunch of these, you begin to understand what’s bothering your visitors. If I’ve got a specific idea that I’m trying to remove from the list, I’ll spend some time watching session recordings.
Sticky heatmap.
A more advanced conversion optimization strategy is running an eye tracking study. Now, this doesn’t work directly with your website, but you can bring people from your website if they’re willing to take a look. And it’s just amazing that this technology exists because eye-tracking studies used to be so hard.
Submit a mockup to a company called Tobii, and they’ll bring 25, 50, 100 people to look at it. They’ll record what the visitors see on the page using laptop cameras. Laptop cameras have such high resolution that we can tell where people’s eyes are looking on the screen.
We can see what people linger on, what ideas they like, what offers they like, and where there are images that stop them on the page. This information is really valuable if you’re trying to critique your page layout.
Gamification: AB Testing
The last thing that I spend time on is AB testing. Because if we’re going to take something to an AB test, as it’s the best data we can collect, we only want to take the best ideas. And it takes quite a bit of work to get AB test results.
Here is an example of one that we did. We worked with a company called Automatic and they had a plug that plugs into your car and connects your phone to your car’s computer. They came out with this new Pro version, but everybody was buying the Lite version.
Why wouldn’t people buy the Pro version? Sure, it’s more expensive, but it’s so much better. Maybe we’re not communicating how much better it is effectively.
We did a “Thank you” page popup survey asking, “What made you choose Lite instead of Pro?” We found out that people didn’t understand the value of the Pro features.
We created a version of the product page was simpler. It was a shorter list of features, and we only highlighted the things that were most different. This is something you should consider any time you’re offering multiple plans or products on a pricing page.
We designed an AB test. One half of the visitors saw the original page, which we call the Control. The other half saw our variation. The result was a 13% increase in conversion rate for our variation. We also saw an increase in revenue per visit because more people were buying the Pro version.
A/B Testing: automatic pro vs lite.
This achieved exactly what we wanted. The data we collected during the AB test was very reliable, because these tests are designed to eliminate as much randomness as possible.
What are the Steps of Conversion Optimization Summary
If you are going to be a successful digital marketer, you are going to be an experimenter. Your ability to use the tools and data of the trade will determine your future in a data-driven marketing economy.
How Often Can I Email my B2B List: Four Lessons Learned
Conversion OptimizationOne of my favorite conversion strategies is the second chance. The second chance only comes when I have a way to continue the conversation; to get someone to come back again and let me make my case again.
There is no better second chance channel than email.
When entrusted with an email address, and permission to continue the conversation, I have one, two, three or more chances to persuade a prospect to reconsider.
In a business-to-business situation — the considered purchase — in which a decision will be made over a period of weeks or months, email is a true friend. And if it is executed with respect, it is a friend to those struggling with a purchase decision.
The question is, how many second chances am I going to take?
Five Emails an Hour
I once got five emails from American Airlines within the space of an hour. Did I unsubscribe? Did I feel spammed? The emails were telling me the status of a flight I was booked on as its departure time and gate changed. The emails were completely relevant to my situation, and were welcome.
If we were to stand by our statement that businesses can send as often as their emails’ relevance allows, we need to understand the dynamics of a high-frequency email campaign.
How Often Can I Email my B2B List: An Email a Day Experiment
The goal of this experiment was to examine the following hypotheses:
a. The List
We chose a selection of 2000 names from my house list. This list consists of contacts made through personal interactions, meetings and consultations. It is primarily a business-to-business email list.
I would call the list a “semi-warm” list having received email from me only quarterly. This list had received emails on January 11 and April 30. The experiment began September 7.
Your list could easily be generated from social media traffic or search engine traffic.
b. The Content
Because of the frequent nature of these emails, it was important that they provide some value and be entertaining. This proved to be a significant challenge.
Each email followed the following formula:
Subject lines included “Are you the victim of the Email Invisibility Ray?,” “Social Media: Marketing from my La-Z-Boy,” and “Why eight-year-olds beat me at Chess.”
3. The Frequency
Emails were sent daily, Tuesday through Friday for two consecutive weeks. Eight emails we sent in all.
High Frequency Email Campaign Test Results
1. Email Performance vs. Social Media
We’ve had relatively good luck using social media to drive traffic to my site. However, in Figure 1, you can see that the email resulted in significant increases in traffic, even outperforming our summer social media experiment.
Figure 1 • Traffic sources overview: email effect on site traffic.
One interesting note is the rise in search engine traffic at the time of the email. This underscores that click-through rate is only a partial measurement of email effectiveness.
2. Increased Conversion Rate
It is probably not surprising that sending email to a targeted list is going to result in more conversions. However, keep in mind that my social media networks are also quite well-targeted.
As expected, both conversions and conversion rates for new subscribers increased. We can also attribute thirteen (13) workshop registrations to this email series, generating almost $1300 in sales.
Just looking at new email subscribers, the conversion rate for our social media experiment were 2.5%. For the period of this email, conversion rates were 7.6%.
Figure 2 • Emails’ Effect on Conversion Rate.
3. Opt-out Rates
This was the metric I was most interested in examining. How would unsubscribe rates change over the course of the experiment?
Figure 3 • Open rate, Click-through rate and Bounce Rate for each drop.
I consider an unsubscribe rate of 1% or less acceptable and expected in any email that asks the reader to take action. So, I got pretty nervous as unsubscribe rates rose to 3.2%, and stayed well above 1%. Over the course of the experiment, 15% of the list unsubscribed.
There are two ways to look at this:
If my goal with this list was primarily to sell, I would consider the 15% loss to be acceptable and even desirable. This is called Shaping your list.
However, my goal is to evangelize conversion and to educate, so the opt-outs represents a pretty significant loss of reach.
From a brand perspective, there were very few negative comments, and many positive ones.
Given the opt-out rates, would I do this again. The answer is a resounding yes.
4. The Effect on Deliverability
The other negative effect that frequent emails can have is an increase in spam reports.
For most service providers, deliverability is the inverse of the bounce rate. If my emails are reported as spam, we would see an increase in bounces. Intuitively, when shaping a list, you expect bounce rates to drop quickly as bouncing addresses are removed from the list.
For our experiment, the bounce rate began at 2.5% but quickly dropped, leveling at an imperceptible 0.06%.
One reader was kind enough to let me know that they had “spammed” my email. I used the site MXToolbox.com to see if my domain had been placed on any black lists. However, it would be our Email Service Provider (ESP) that took the hit if spam was reported. This is one big value of an ESP. They keep themselves – and you – off of black lists.
Figure 4 • Unsubscribe Rates for the Email Series.
Another measure of reader interest is open rates.
Email service providers count the number of times a special image is downloaded to establish open rates. Since many people have images turned off in their email client, the open rate is not an accurate measure of actual opens.
However, I would interpret a steady drop in open rates as a sign that the list is becoming fatigued with my communications. Open rate can also be a good indicator of the quality of your subject line.
Open rates were relatively flat, dropping on Fridays.
Overall, I believe that few of my readers reported these emails as spam.
I attribute this positive outcome to the non-promotional nature of the copy, even though the emails were clearly promoting our email list, workshop and webinar.
How Often Can I Email my B2B List Conclusions
With some simple analytics in place, we can pretty easily establish the ideal frequency of our email campaigns. Based on these results, we should be sending email more frequently. You will probably come to the same result. However, we tested a certain kind of email with this experiment; an email that is informational and entertaining as well as promotional. This style of email requires a bit more work and creativity on our part.
The payoff is quite clear.
Email is a more effective channel in a B2B sale than is social media. It is also a great way to get more out of your search engine and advertising traffic. When you get an email address, you get a second chance at the sale. And a third, fourth and fifth chance.
For the complete content of the emails sent during this experiment, and the results of some split tests conducted, visit.
To the Bat Cave! Your Own Secret Conversion Marketing Laboratory
Conversion OptimizationSometimes it’s better to ask forgiveness than permission
Warning: this information will make you a more successful marketer, but may also put your immediate job in jeopardy.
To be a true hero, you must have two things:
Unfortunately for those of us in marketing, our nemesis is often the organization in which we work; that Dilbert inspired, plodding structure full of people that think they know how to market. Such a beast is often resistant to our most powerful weapons, such as positive results.
The best way to defeat such a daunting foe is through patience and stealth. As marketers, we must build our strength, our knowledge and our skills.
How to set up your conversion marketing laboratory.
Your Secret Conversion Marketing Laboratory
I propose that you consider building your own secret conversion marketing laboratory, your own Xanadu. This is the place you go to explore new marketing strategies and ask questions that others may not have the guts to ask.
Questions like:
These are the questions that take time to sell internally, especially when you don’t have the data. These are the concepts that IT is designed to thwart. It’s time to unshackle yourself. Build your own conversion laboratory.
Rules of Engagement
Now, as heroes, we want to do good in the world. This means doing no harm to our organization’s brand. We don’t want to work against our organizations already plodding attempts to communicate.
We want to minimize cost – most of us aren’t Bruce Wayne – and maximize automation. This will make our time in the lab most productive.
I cover all of the guidelines in my Search Engine Land column Setting Up Your Own Conversion Lab, Part 1.
Why Do We Need A Conversion Marketing Laboratory?
Because conversion marketing is a momentum game. It requires trying things to find out what works best. It requires rapid question-test-analyze-question cycles. And sometimes we have to test unintuitive assumptions to understand our audience.
Without the lab, there are blocks to momentum.
IT has their gatekeepers that slow our testing cycles. Management wonders why we aren’t writing a press release or blog post.
While most marketing departments think they know best, our lab lets our visitors tell us what they want. This is powerful knowledge. There are some big wins to be found in the lab, especially at the beginning.
The Secret Conversion Laboratory
Consistent measurement trumps accurate measurement. Conversion marketing means making decisions based on data. Analytics provide that data.
We aren’t interested in an analytics implementation that is accurate down to the visitor. Instead, we want analytics that are sufficiently correlated to reality.
This is scientist-speak for “when things change, our measurement changes by about the same amount.” When more people visit, our metric “visits” goes up by about the same percentage. It mirrors reality.
Don’t waste your precious time trying to get accuracy in measurement. Good enough is good enough.
Most analytics systems are easy to set up, or are competently integrated into most of the online services you’ll be using in your lab.
Equipment cost must be “under the expense line”. The secret lab is, by design, not going to be a budget line item. That defeats the purpose.
Instead, you need to select tools that are free or cheap enough to purchase and implement without going through the budget process. They need to be expensible.
Avoid IT obstacles. The equipment you use in your conversion lab must not require IT resources to set up and use. IT is too often a bottleneck.
We will be selecting tools that almost any marketer can use. With a little practice and some training videos, you will be able to implement almost any test you can imagine.
It should be highly automated. We must get our marketing duties done with excellence, so our conversion lab can’t take a large chunk of our precious time. If you’re off in the lab for hours at a time, people will begin to wonder. It draws attention.
We will be looking for tools that automate the lab, and solutions that collect and aggregate data for us.
Your efforts should not harm the live web site. Our goal is to become better at marketing for our companies. As such, we should do no harm. Our lab should not:
Basically, we want to do small tests, learning things we can use to help the company sell more and dominate online.
Beakers, Bunsen Burners and Mass Spectrometers
We are fortunate to have many of the tools needed in our lab available for free or at low cost.
You will need tools to:
Cape and tights are not required
It may be tempting the done a hero’s uniform once you begin to feel the power of what you learn in your lab. Honestly, It’s best to stay under the radar.
Let us know which tools you find in your lab in the comments, and please share any interesting results you get from your experiments.
Originally published on the Search Engine Journal
Can AI Marketing Tools Increase Your Website’s Conversion Rates?
Artificial IntelligenceDo you know how machine learning is impacting conversion rate optimization for marketers? We all know what the acronym “AI” stands for: “As If”. Data scientists are telling us that by using AI, they’ll will be able to create a predictive model of the visitors to your website that will tell you exactly who is ready to buy.
I say, “As if.”
We may marvel that such things can be done, but we also recognize that these things require a great deal of data and the skills of some serious brainiacs to get a machine to tell us something we don’t already know.
The truth is, you are probably already using “AI”, or more accurately, machine learning in your marketing. It’s hiding in the tools we use, like monsters under our bed. Machine learning and the more sciencey-sounding AI will change the way you take products to market, but your human mind will still be needed and loved.
Unless you resist – “as if.”
Augmenting Our Brains: AI-powered conversion optimization
We say things like, “If you give me a budget, I’ll generate six times that amount in revenue.” This is is like saying, “If you put a chicken foot under your pillow, you will find true love.” As if.
But this is what we do, and the data on which we base our predictions is often no more valid than the layout of tea leaves at the bottom of a cup. Our brains are wired to find patterns in anything, even when a pattern isn’t really there.
If I came into your office and said, “The last three leads we generated were all visiting the website using a Firefox browser,” your brain would jump to the conclusion, “If I can get more Firefox users to visit our website, I’ll generate tons of leads.”
Do AI marketing tools impact your website conversion rates?
Purveyors of AI, or more accurately Machine Learning (ML), would tell us that the machine doesn’t make mistakes like this. Our 100% genuine intelligence just doesn’t stack up to their Artificial Intelligence.
The problem is that machines will make exactly the same mistake if we don’t give them lots of data.
Just as machines need data, we know that we need more data before we start an ad campaign targeting Firefox users. We’ll ask our analytics person to pull together all website visits for the last year, and calculate the conversion rate for each. This increases the size of our dataset from three to many.
In this scenario, we “wasted” an hour of our data scientist’s time, an hour of our time, and another twenty minutes explaining to our boss why we were so unproductive today.
“What if,” the AI crowd says, “you could get a machine to sort through your data looking for clues and figuring out who’s more likely buy. You don’t have to waste your time. Let the machine do it.”
This is an exciting proposition. The machine wouldn’t just look at the browser. It would look at the time of day, day of week, and week of the year that visitors converted. It could consider the device being used, screen size and operating system. It could add in the source of the visit, the number of times a visitor has been to the website, and whether the visitor has bought before.
After crunching through all of your analytics data, the machine would give you a percentage chance that the next visitor to your website will convert. And here comes a person with a Safari browser on a Mac computer at 3:30pm EST on a warm Tuesday afternoon who’s never been to the site before.
The machine might spit out, “There is a 51% chance this person will complete the lead form.” Actually, the machine will just say, “0.51”. Machines are so boring.
It’s amazing that a machine can so accurately predict a human being’s behavior. This is incredible.
This is “the rub” with machine learning. The machine can’t tell us what to do with the data it gives us. There are systems that will tell us if a visitor is “at the top of the funnel” or “in the consideration phase.” Still, what do we do with that? A price-sensitive buyer may want to see a discount when “at the top” of their purchase process. A relational buyer may not care about discounts until they’re “at the bottom,” ready to buy.
The machine won’t tell us, “Target Internet Explorer visitors coming late at night on a Windows computer during the springtime months with a picture of a cat.” It spits out the probability for each visit: “0.51, 0.34, 0.71, 0.92”.
Wait! A 92% probability? Is that important!? Well, no. They’re probably going to buy no matter what we do. “As if.”
AI-Driven Results
Scoring customers in a customer relationship management (CRM) platform has required that marketers hand-code the algorithm. We decide which actions indicate that a prospect is moving closer to buying. We decide how to value each action. It can work, but it isn’t rocket science – or AI.
Alternatively, we can dump sales data into a machine learning algorithm and let it calculate the probability that each prospect will turn into a customer. The sales force can focus on those high-probability clients and disregard the low-scoring leads. It’s using past performance to predict the future, and should be more accurate than arbitrary assignment of values to actions.
This is how machine learning is entering your life as a marketer.
AI Conversion Rate Optimization: Can AI Marketing Tools Increase Your Website’s Conversion Rates?
It’s not an easy problem to solve. There are a lot of variables to crunch and it has to be done quickly. This is a prime area for AI.
Mailchimp launched a similar tool to add product suggestions to the emails of its eCommerce clients. Every time you send an email to someone, Mailchimp will include a few product suggestions at the bottom of the email. The machine learning algorithm will compute the probability that one or more products will appeal to a subscriber, based on the behavior of all email recipients. Those products with the highest probability get added to the email. This prompts the visitor to buy.
As if.
It’s hard to know how well the machine has learned what your visitors buy collectively. This is the limitation of AI. We can’t really see what is inside the box. All we get is a number.
If you implement a suggestion engine on your website, we recommend running an A/B test to measure its effectiveness. This is done by adding the “Also bought” suggestions for half the visitors and hiding it for the other half. This will give us some conclusive data about how the suggestion AI is performing. Is it increasing the order size on average, or reducing it?
Eliminate Risk Upfront and Bump Your Lead Conversion Rate
Lead GenerationEven in B2B lead generation, you can “reverse” the perceived risk of completing a form. Risk reversal tactics are an important part of conversion rate optimization.
There are lots of reasons that someone wouldn’t fill out your lead form, even if your white paper, podcast or webinar are free.
Well, even in business-to-business lead generation, you can “reverse” the perceived risk of completing a form.
“Hey, you. Yes, you. Come here. You wanna white paper? Well, I have a deal for you. All you gotta do is give me your contact info. Hey, you got nothin’ to worry about. I’ll take real good care of the information. Honest.”
I’ve been studying the landing page techniques of what I call “The Bad Boys of Conversion.” These marketers, often called “affiliate marketers” or “infopreneurs” get high conversion rates for everything from get-rich-quick systems to health supplements.
My goal is to understand how the well-tested tactics they use could be applied to more conservative business sites.
One of the most important techniques used by these marketers is “risk reversal.” Money-back guarantees are perhaps the most common form of risk reversal. Are these tactics useful for business-to-business lead generation?
Absolutely, but a different kind of risk reversal is in order.
Risk Reversal Tactics do work on Business professionals
Business professionals respond to the same incentives that consumers do. However, consumers are risking their money, while businesspeople may be risking their reputation, and their decisions could affect their jobs. This makes business prospects more risk averse.
Thus, using risk reversal strategies becomes even more important in a B2B transaction.
You can significantly increase both your conversion rates and the quality of your prospect list by using some of these risk reversal tactics on your lead generation pages.
7 Risk Reversal Tactics For B2B Lead Generation
Remove the risk up front
Make your white paper, webinar, or video free. Before you say, “Lead generation content is always free,” keep in mind that your prospect is “paying” for the content with their contact information. To them, it isn’t exactly free, and the decision to give you their contact information is similar to the decision to spend money.
Here are some ways to remove risk up front.
1.Don’t ask for contact information.
If you have a piece of content that really nails your value and generates qualified calls give the content away. There is no risk to the prospect, and broader readership can be expected.
Removing risky fields can also lower the “cost.” Fields that increase the cost of lead generation content include address, phone and many qualifying fields. Do you need their address? Do you need to know their marketing budget, really? These risky fields can decrease your conversion rates.
Ask for only the information that you need to qualify and contact them, and eliminate all the rest.
2.Make your forms optional
This allows the prospect to select the level of risk they are willing to accept.
Completed forms will indicate better qualified prospects, prospects that already have a level of trust with your brand. Partially completed forms, less so. Empty form submissions indicate prospects that you may not want to waste time on anyway.
3.Emphasize respect of privacy
On your landing page, best place to put risk reversal is near the submit button (but it shouldn’t ever say “submit”).
The simple phrase, “We respect your privacy” communicates two things: first, that you have a privacy policy, and second, that privacy is important, and you’re not likely to abuse this information, or give it to inbox-clogging spammers.
It is common to make the word “privacy” a link to your privacy policy, but that may not be wise. In his excellent book “Always Be Testing: The Complete Guide to Google website Optimizer”, Bryan Eisenberg recommends testing this. The link takes an interested visitor away from the landing page, and this may decrease conversion rates.
4.Ask for opt-in, not opt-out
If you’re asking a prospect to sign up for email communication, you should explicitly ask for permission. You may feel that this is implied when someone provides their information to you, but any uncertainty increases risk.
Tell the prospect can they can expect and let them opt-in by checking a box. This is an opportunity to sell your email communication.
“Check this box for occasional tips and best practices by email on widget use.”
or
“Subscribe to “Widget News,” the authoritative email newsletter on widgets and their application.”
Allow opt-out at any time. Let prospects who are signing up for email communication know that you will be offering opportunities to opt-out of the email with each communication.
“You can unsubscribe at any time.”
You can use any specifics from your privacy policy, such as “we do not tolerate spam,” and “Your information is digitally encrypted.”
5.Spell it out
Transparency is a great risk reverser. Tell them on the page what will happen when they click the “Get your report” button.
“You will be taken to a page that contains a link to the report. You will also receive an email asking you to confirm your email address. We don’t tolerate spam. Based on your input, a respectful sales person may call, and this is your opportunity to ask any questions about widget use in your organization.”
You should also tell them what they won’t get. “No one will call” is a common risk reversal phrase. Of course, this limits what you can do with your house list, and you should be sure to adhere to any limitations you promise.
6.Add risk and then remove it
On the scales of a decision you have risk balanced against value. If your landing page does a good job of building the value of your offer, you will find prospects more willing to take risk, and complete the form.
One great way to communicate value is by setting a price. Consider charging for your content. Then offer a money-back guarantee to reverse risk. No questions asked.
Does this sound like raising the prices in a store just so the owner can run a sale? Yes it does. But our goal is to communicate value—not to boost our profit.
This strategy will reduce conversion rates, but it will deliver highly-qualified prospects.
For example, charging for a product trial—a common lead generation offer—will often ultimately deliver more paying customers than a free trial, even though it reduces the number of people who try the product. Since we’re not really trying to make money on the offer, we won’t be concerned about those who request a refund just to get free content. We should all have content worth stealing.
7.Always deliver on your promises
You must be able to support your risk reversal claims. You must enforce your privacy policy internally. Don’t email prospects if they opt-out, and don’t let a salesperson call someone on your list if you promised you wouldn’t.
Above all, deliver the content that they “bought” with their contact info and their attention.
I use a unique email address for every site I provide personal information. When I get spam, I check the address to which the email was sent, and that tells me who’s been stealing my contact information. With services such as OtherInBox, more professionals are using this technique. It’s more and more likely that you will get caught, and your brand will suffer.
You don’t want to join the “bad boy” marketers, do you?
In general, you can reverse risk by
This article was first published in my Search Engine Land column.
7 Simple & Effective Ways to Improve Your Email Open Rates
Conversion OptimizationEmail is one of the highest converting marketing channels and few activities drive a higher ROI than building an email list. But having 50,000 email subscribers doesn’t really matter if you can’t get any of them to open your emails.
We’ll show you seven relatively simple (yet criminally underused) strategies that are GUARANTEED to improve your email open rates. By the end of this article, you will have an actionable plan for significantly boosting the ROI of every email you send.
7 Surefire Ways to Improve Your Email Open Rates.
1. The Best Way to Improve Your Email Open Rate: Work on Your Subject Lines
Your email is desperately fighting for attention in your prospect’s inbox. Your job is to differentiate yourself within that inbox, and the only way to do that is with your email subject line. But it’s not just about standing out. What is most important to boost those email open rates is that your subject line is relevant to your audience.
Let’s take a look at some examples from my inbox:
Work on your subject lines if you want to improve your email open rates.
Remember that the goal of a subject line is not to get just anyone to open the email. You have a specific audience you’ve built, and it’s important to use the types of headlines that will resonate with that audience.
Want more examples? Get inspired by these 165 great email subject lines.
2. Match Your Lead Magnet To Your Content To Your Product
There’s a lot of content out there about how to build an email list. But one of the biggest mistakes I see businesses and bloggers make is thinking too short-term with their lead magnets and other lead acquisition strategies.
Email marketing is a scalable form of relationship marketing. It’s meant to be a long term pursuit, and accordingly, it’s important that what you are talking about remains consistent throughout that process.
If you attract someone to your list via an ebook on increasing web traffic, then start emailing them about how to run an ecommerce store, and then try to sell them a product on increasing productivity, you won’t have much success.
Every part of your sales funnel should be designed to appeal to the same customer profile.
When someone signs up to your list to hear about a given topic, the emails they receive should also be on that topic. By aligning your lead magnets, emails, and products, you can significantly improve your conversion rates across the board.
3. Split Test Until You Find The Optimal Send Time
One of the easiest ways to increase your open rate is to send your emails at the right time.
How do you find the right time? Two words: AB testing
This requires a little bit of intentionality upfront, but it’s so easy to do, and it can have a MASSIVE impact on the reach of your content.
Here’s one way to do it.
Schedule your next 7 emails to go out on different days of the week. For each email, break your subscribers into 4 identical groups and send to each group at a different time of the day.
This will give you a total of 28 different send times to compare against each other. If you have a smaller list, you might need to run this experiment several times in order to get statistically significant results, but what you should start to see is that certain days and times will consistently perform better than others.
By sending at these optimal times, you can significantly boost your open rates.
4. Delete Prospects that Never Open your Emails
Sometimes email subscribers just have to go. It’s hard to part with those emails you spent so much effort collecting but evidently they are not responding to your standard email campaigns. Cleanse your email list regularly as it will also help you improve deliverability. Gmail and Yahoo emails will pick up on this and send your email to their junk folder where nobody can see it, much less open it. And this makes a huge difference to your email open rates!
If you don’t want to get rid of them, split your audience and send them a very different kind of email. A farewell email. Help them move on or open your email to stay.
5. Deliverability is Key to Improve Your Email Open Rates: Resend To Unopens
This is the easiest strategy we’ll discuss today. Even when you do everything right, a lot of your emails will never even be seen.
Thanks to Gmail’s inbox categorization. A lot of emails now fall under the promos tab. There are some things you can do to try and get in the primary inbox, but that’s a bit too complicated for today’s post, so instead, I’m sharing a simple trick that anyone can use to instantly increase opens, often by an additional 25%.
Simply resend the email to anyone who didn’t open it up the first time. I like to wait around 3 days to resend, but you can experiment for yourself and see what happens.
One email is nice, but two could get the job done.
In the example above, the original email performed pretty well, so the resend was a bit less powerful than usual. But even if your list is as small as 5,000 subscribers, that additional 7% in total opens means nearly 400 more people are seeing your message.
6. Be Authentic: The Email Campaign Subscribers Look Forward to Opening
One of the first things people notice when signing up to a new email list is tone.
A lot of marketers and business owners nowadays are just running email marketing because they’ve been told to. They are following a template and trying to sell products. And while there is nothing wrong with that, it’s very, VERY obvious to your readers.
The most successful email marketing campaigns often come from the biggest personalities. When you sign up for their list, you aren’t introduced to a sales pitch. You are introduced to a personality.
You read personal stories about their background and experiences. You read about their fears and failures and ultimately, their successes.
Pretty soon, you feel like you know this person sending you these emails.
That’s the power of being personal.
And while I can’t teach you to have a personality, I can give you some pointers:
Being personal is not hard, but it does require you to get out of “marketer mode” from time to time.
7. Authenticate your Email Sender: SPF, DKIM, DMARC
If your emails are not reaching the inbox of your subscribers then your email will never be opened. If you want to improve your email open rates, prove to the servers – particularly big corporation servers – and subscribers that it is really you who is sending out the emails.
Here’s what you’ll need to do to get your email authentication going.
Setup Sender Policy Framework (SPF) Authentication.
Setup DKIM Authentication (DomainKeys Identified Mail) to help detect forged addresses in an email.
Enable DMARC protocol to protect your email from spoofing.
Conclusion: Increase Your Email Open Rates
Let’s review:
By implementing these strategies, you can immediately increase the open rates of your email campaigns.
Easy.
Before you go, we’d love to hear from you. Have you tried any of these strategies or do you have any tips of your own to add?
Let us know in the comments!
Reduce Bounce Rates: Ready to Fix Your Conversion Problem?
Conversion Optimization, Landing Page Optimization, Web AnalyticsA bounce is any visit for which the visitor only looks at one page and does not interact with it. This sounds truly unfair as someone may spend minutes on your blog post or landing page, and still be counted as a bounce.
A visitor bounces when they don’t find anything close to what they were looking for when they visit your site. Either you’re attracting the wrong visitors or you don’t know why they are visiting.
Bounce is the most extreme form of conversion problem. High bounce rates are an indication that you are throwing good marketing dollars down the tubes. Whatever you’re spending to get traffic to your site is being wasted.
How to Reduce Bounce Rates or the heartbreak of “bounce”
Boing!
That’s the sound of someone finding your site, but not finding what they wanted ON your site.
Boing!
That’s the sound of website content that doesn’t match your marketing.
Boing!
That’s the sound of a website that talks about the company instead of the visitors’ problems.
Bounces Aren’t Helpful to Businesses
What are some strategies to reduce bounce rate?
This is a common question, and requires an understanding of the definitions of bounce rate.
The bounce rate is a bit slippery and requires some examination. The intention of measuring the bounce rate is to figure out how many of your visitors are leaving almost immediately after arriving at your site. This metric provides for a lot of error in interpretation.
This is rarely the case. A more accurate explanation is that your site doesn’t look the way your visitors expect it to look. Understanding what your visitors expect is the way to reduce bounce rates.
Instead, there are usually some more valid reasons for your high bounce rate. Here are the things digital marketing and conversion experts examine when confronted with uncomfortably high bounce rates.
1. You’re measuring it wrong
How you measure your bounce rate can give you very different insights. For example, blogs often have high bounce rates. Does this mean that visitors don’t like the blog?
Many analytics packages measure a bounce as a visit, or session, that includes only one page on your site. Visitors who take the time to read an entire article would be considered a “bounce” if they then left, even though they are clearly engaged.
We set a timer for our blog traffic, so that any visitor who sticks around for 15 seconds or more is not considered a bounce. You can set a timer to the amount of time you consider appropriate.
2. How to Reduce Bounce Rates: Diagnose Technical Difficulties
We are fond of saying that you don’t have one website, you have ten or twenty or thirty. Each device, each browser, each screen-size delivers a different experience to the visitor. If your website is broken on one of the devices popular with your visitors, you will see a bump in overall bounce rate.
If your pages load slowly, especially on mobile devices, you can expect a higher bounce rate.
Broken internal links and 404 pages are also cause for bounce.
If your page breaks out in a chorus of Also Sprach Zarathustra when the page loads, you may enjoy a higher bounce rate.
How to diagnose device-related technical problems
Your analytics package will track the kind of device your visitors are coming on.
Is there a problem with this site when viewed with the Safari (in app) browser?
The Google Analytics report Audience > Technology > Browser & OS shows that there may be a technical issue with Safari visitors coming from within an app. This may also reflect visitors coming from mobile ads, and they may simply be lower quality. See below.
With Google Analytics Audience > Mobile > Devices report, we see mobile devices specifically. The Apple iPhone has an above-average bounce rate, and we should probably do some testing there, especially since it’s the bulk of our mobile traffic.
With an above average bounce rate, visitors on an Apple iPhone may be seeing a technical problem.
3. Good Traffic Quality Helps Reduce Bounce Rates
If you are getting the wrong visitors, you will have a high bounce rate.
Remember StumbleUpon? Getting your site featured on the internet discovery site often meant a flood of new visitors to your site… and a crash in your conversion rate. Stumble traffic was not qualified, they were just curious.
Your bounce rate is a great measure of the quality of your traffic. Low quality traffic bounces because:
We take a closer look at the source of traffic to diagnose a traffic quality problem using Google Analytics Acquisition > All Traffic > Channels report.
Direct is one of our biggest traffic sources and brings in one of the two highest bounce rates.
Here we can see that traffic coming from social media and those visitors coming “Direct-ly” have a high bounce rate.
If you are driving a lot of visitors to your home page, you may want to consider presenting them with links to more relevant content. As Tim Ash says, “The job of the home page is to get people off of the home page.” He didn’t mean by bouncing.
With regard to social media, we may have a problem with broken promises.
4. Broken Promises Lead to Conversion Problems
Do your entry pages consider the source of visits?
If your traffic is clicking on an ad that promises 20% off on a specific propane grill, and they’re directed to your home page, you’ve broken a promise.
You might think that they will search your site for the deal. You might even think they’ll search your home page for the deal. You’re wrong. Many will jump.
Every ad, every email invitation, every referral link is a promise you make to your visitor. If they don’t come to a page that lives up to the promise, they are likely to bounce.
This Ad takes the visitors to a page that is almost designed to disappoint.
Looking at your ads on a page-by-page basis is necessary to diagnose and correct this kind of bounce rate problem.
5. Vague Value Propositions don’t help reduce bounce rates
Ultimately, if you’re not communicating your value proposition to your visitors clearly, you are going to enjoy a monstrous bounce rate.
21 Quick and Easy CRO Copywriting Hacks
Keep these proven copywriting hacks in mind to make your copy convert.
"*" indicates required fields
Your value proposition typically does not address your company or your products. It should be targeted at your visitor, why they are there, and why they should stick around.
Each page has its own value proposition. Your business may have a powerful value proposition, but each page should stand on its own.
A contact page should talk about what will happen after you complete the form. Who will contact you? How long will it take? Will they try to sell you something?
A landing page should clearly state that you are in the right place and provide reasons for you to stay and read on.
This landing page delivered a strong value proposition in above the fold. See the full case study and video.
A home page should help you find your way into the site. Most home pages are treated like highway billboards. No wonder people just drive on by.
6. Finally, Is Your Stubborn Bounce Rate Mocking You? [AUDIO]
Most of us, at one time or another, have not only been frustrated with our bounce rate, but completely puzzled about what to do about it.
In this podcast, I will guide you toward a more accurate bounce rate. I will walk you through the steps to use The Timer Listener to time the visit duration of each of your visitors and get those percentages down where they belong.
Subscribe to Podcast
Richard Strauss: Also Sprach Zarathustra by Kevin MacLeod is licensed under a Creative Commons Attribution License.
Brian Massey Marketing Land article, Using A Google Tag Manager Listener To Get Your Real Bounce Rate