I have a question for you.

How can you know if you’ve optimized your landing page or product page to its fullest extent?

How can you know if you’re leaving revenue on the table?

The short answer is: You A/B Test it.

What’s A/B testing?

Well A/B testing is the experimentation that you do to determine which version of the same asset is better.

It’s randomized experimentation where a sample of visitors or users are shown a variant while another sample showed another.

It’s a competition where you bet A against B to see who wins based on your evaluation criteria, be it clicks, views, lead generation, etc.

And you can A/B test basically anything on your website; you can A/B test your copy, your CTA, your light mode vs. dark mode, anything that comes to your mind really.

Split testing

Now you might hear the name split testing when you’re researching for A/B testing.

Yes, Split testing is A/B testing but it differs in the way it works.

A/B test will show both of the variants on the same URL.

So if you’re running an A/B test, then it’s going to be on a single URL like this figpii.com/a-b-testing

But if you’re doing it as a split test URL then you’ll have 2 URLs instead of one

Figpii.com/a and Figpii.com/b

That’s for A/B testing, but there is also A/A testing.

A/A testing

It sounds counter-intuitive; why would I be interested in betting the same thing against itself? 

I get what you’re thinking but trust me, A/A testing has its place in digital marketing.

It’s actually a good way to make sure that your A/B tests are accurate because if your data is skewed by something, that might ruin your whole experiment. 

So now that you know what’s A/B testing let’s dive deeper into how do you do those kinds of tests?

How A/B Testing Works

In a nutshell, to do an A/B test, you need to first determine what are you going to be testing.

Everything is testable in this modern age; you can test your hero section for an example.

You can test your copy when sending your users through a funnel page.

Or you can test different images and other kinds of media to determine which one of them is the better performing! 

Let me walk you through the process of creating your first A/B test using FigPii

1. Log in to FigPii. 

2. The FigPii code will appear at the bottom section of the dashboard3. Click on “copy code”4. Paste the FigPii Tracking Code into the <head> section of your website.

3. Click on “copy code.”

4. Paste the FigPii Tracking Code into the <head> section of your website.

FigPii integrates with major CMS and e-commerce websites; check our full list of integrations right here

5. Check your FigPii dashboard to verify the installation. When the tracking code is installed on your site, the tracking indicator will show you that the code is active.

Once you have added the FigPii Tracking Code to your site, you will need to wait for about an hour to check if it is installed correctly. This usually happens the moment your site is loaded with the FigPii tracking code installed in. But there can be a delay for up to an hour before it shows as “Active”.  

6. log in to FIgPii’s dashboard, and go over the A/B testing section from the side menu.

7. Press on Create New Experiment

8. Choose your A/B test name

9. Choose whether your experiment is an A/B test 

10. Choose whether you want to test just a one page or multiple pages 

11. You can choose specific parameters to help you design an accurate experiment so you can test things like:

  • Which copy converts more visitors to customers?
  • Which CTA is getting more clicks 
  • Did changing these sections make any difference?
  • And much more 

 12. Add your first variation, then repeat the step above again to add more variations to the experiment.

13. Create your goals. You can create one or more goals to judge the success of your experiment.

FigPii will use the primary goal to determine whether your experiment “wins” or “loses.” Secondary goals can help you measure other metrics to judge the success of the experiment.

Examples of primary goals include reaching the “order confirmation” page on an e-commerce website or reaching a thank you page after filling out a contact us form. Secondary goals include reaching a different page within the funnel on an e-commerce site or filling specific fields on a contact form.

13. Refine your experiment using our advanced options to make your experiments more impactful.

And that’s it, and you’re done! You now have an A/B testing experiment. Let it run for a couple of days and then check the results and which of your variants was concluded a winner.

Now you might be starting to think, okay, but how does an A/B test help me achieve more?

Why You Should A/B Test

A/B testing should be a routine task for you as a digital or growth marketer.

Simply put, the more experiments you run, the more you know about what works and what doesn’t.

You might have a bias to a copy you wrote or a visual you liked, but that doesn’t mean it’ll be the same for your customers or visitors.

They might like another copy or visual, for that matter! 

And it’s not just your copy or visuals; you need to test everything on your website to make sure that you get the highest conversion rate possible.

I worked at a company before and they had a landing page for downloading an e-book, it was a lead magnet.

And I was debating that the dark mode page would convert much better than a light mode one because the dark mode is the meta right now.

And we designed the page on GetResponse with 2 variants, one is dark blue and the other is white.

We let it run for a couple of days, it got nice traffic since we were pushing it on social media and email.

And the results came in, and I was wrong! 

The light mode page actually converted more leads per visitor than the dark mode one, I think it was a 15% difference.

And so I was shocked, don’t people like dark mode?

And from that point on, if I have an idea, I A/B test it to check the validity of the idea to eliminate the bias.

My experience is quite similar to big brands’ A/B testing results as well.

Hubspot did an A/B test which you can read about right here.

In a nutshell, they did A/B tests on mobile and came to the conclusion that they can generate up to 1300 more leads generated per blog if they changed their CTA.

Let me give you some examples of where you’ll need to A/B test the most.

  • Content

If you want to write better content that engages with your audience, you need to do frequent A/B tests.

You’ll have a bias towards your copy and it might not be the ideal copy for your customer.

If you want to write better content, know what your audience wants, then A/B testing is your key.

  • Less Bounce, More Conversions

One of the important reasons that we do A/B testing is to turn hypotheses into facts.

So if you think that adding a CTA to the bottom of your email will generate more conversions test it, run your A/B test, and see the results.

If it causes more people to not open / not respond then you know this isn’t the right move.

But if it yields good results, then you make it the default.

  • Better Analysis

We all know and love Google Analytics and other awesome analytics tools, right?

But they don’t give you that many great insights into what your customers think right?

That’s why A/B testing is important for you, it empowers you to test multiple assets at the same time, you don’t have to have the changes fully deployed to test.

  • Minimizes risk percentage

Imagine that you were tasked with changing the home page whole design, and you have your new design ready to deploy.

And you deploy it, you’re feeling strong about your new design,

But the days go by and your CTR is going down, not up.

You start panicking, what did you do wrong?

Do you really want this happening to you?

  • Reduce Shopping Cart Abandonment

This one is a classic in every eCommerce store’s marketing book.

There is a lot that you can do to try and minimize your cart abandonment rates.

SO you have a lot of playgrounds that can improve your cart abandonment rates without you even breaking a sweat.

You can test:

  • The time in which you send your email
  • Subject Line
  • CTA
  • The copy itself
  • Personalization
  • And much more

A/B Testing Process

We’ve covered how to run an A/B test, but what’s the process of running A/B tests is like?

  • Do Some Research 

A/B testing is a plan, a plan that you count on to improve your marketing plan, so first of all, you want to do some research.

What do you expect from this A/B test? What’s the conclusion that you’re looking for?

What’re the variables that you’ll be testing, you don’t want to run multiple A/B tests that can influence each other so your data doesn’t become corrupted.

  • What’s your hypothesis?

When you’re doing A/B testing, you have formulated some theory in mind that you’re now trying to prove or debunk.

Document your hypothesis so that you can go back to it later down the line.

You never know, it might turn out to be a great reference point in the future.

If it works you might have more confidence running with it in the future since it worked for you previously.

And if it failed that great, you need to dig deep into why it failed and what can you learn from that failure.

How many A/B to run?

How many tests do you run for your business per week? 

What’s a golden number or A/B test that you need to run anyways?

The idea behind this question is simple: there’s an expectation from people involved in A/B testing that we need to run a decent amount of A/B tests for every given week. 

There’s nothing wrong with having such expectations and sometimes breaking them even makes sense.

But there are also times when it doesn’t make sense to do so, especially if you have not yet reached any significant improvements based on your initial or current experiments.

Anyone who has ever run some past experiments prior to getting started with ABT would know how easy it is to be enthusiastic about launching new A/B tests. 

For example, you may want to test-launching new landing pages or adding a new blue button. We can’t really blame ourselves for doing so as it is extremely exciting.

But that same enthusiasm might create another problem: we are not going to stick to our plan in terms of learning from what our experiments have taught us already and stopped running any new A/B tests just because we ran out of time (or didn’t feel like running them anymore). 

Or worse yet, we never even learn why some of those initial A/B tests failed (assuming they did).

What does the research say?  

Although there isn’t much research on this topic, most findings seem to indicate that testing 2-3 A/B tests per week is optimal.

This research might not be very helpful for those of us who run 2-3 A/B tests on a weekly basis anyway, what it does do is give us evidence that we are probably doing just fine.

A/B Tests Success Rate.

You might be surprised by the success percentage of A/B tests.

I thought that my tests were flawed when my failure rate was high.

But I researched the subject and found out that according to UX planet, A/B tests have a success rate between 12%-15%.

However, depending on the number of users and conversion data when you start a test, this success rate can drop below 10%.

This means that A/B tests are a very effective method to see if an idea can work or not. 

But like any other hypothesis, some of them will be true and others will be false.

So even though I got bad results with my first A/B tests I kept trying until my 5th attempt where things started to change for me. 

But let’s look at how these experiments became successful in my case.

The best advice I can give you is to try out lots of different ideas and don’t stop implementing changes just because your last test didn’t bring good results. 

How Long Should You Run an A/B Test

You’ve identified your target audience. 

You’ve designed, coded, and tested your landing page. 

Now it’s time to come up with an A/B test and a hypothesis for how you can improve the conversion rate from visitors to leads. 

How long do you run the experiment? 

Is there such a thing as too long? 

When is it time to declare one variation as more successful than another and move on?

According to Kissmetrics a week or two should be plenty of time but it depends on how complex the test requires you to set everything up.

And According to Invesp, the maximum duration should be around four weeks.

There’s a very good reason for this. 

This time frame is long enough to allow sufficient data to be collected from users while also minimizing the impact of site-specific external factors that may affect your results.

Which Are the Best Elements to A/B Test?

Which Are the Best Elements to A/B Test? 

There are plenty of elements that you can test in A/B testing, but you need to focus on the most important ones. We

Headline

The first thing that users look at in your website is your headline. therefore it needs to be interesting and eye-catching.

 It’s very important to test different headlines for the same page. In some cases, you might find that one headline works better than others with a certain audience.

and subheadlines as well. It also helps if your headlines are clear and easy to understand without reading the whole text of it.

Call-to-Action (CTA)

The most important element of your website is the “call-to-action” or CTA button, which is why it needs to be eye-catching.

It doesn’t necessarily mean that you need a big picture of a button but make sure its text and font are clear enough to make users click on it.

You can test multiple things like:

  • Button’s copy
  • Color
  • Position
  • Size 

Freemium vs. Free Trial

What you want to test here is actually pretty simple, you want to see what works best for your audience.

Set the barrier low so more people can signup and use your product.

You can test a couple of hypotheses in this regard.

  • Trial vs. Freemium 
  • The duration of the trial
  • Lifecycle Emails

Lifecycle emails are short messages or offers that you send to users as they go through the stages in their purchase cycle. 

You can have a different message for each stage and test what works best with your audience.

For every step in your funnel, you can test multiple copies to try and achieve the best open to conversion rate! 

Pricing Page

The pricing page is actually pretty simple, but most people fail at this point. 

What seems like the most obvious thing to test is, in fact, not. 

You want to test the combination of your features and price.

But don’t just test your price, you want to make sure that the customers won’t feel like they are being ripped off. 

You can also change some of the elements on this page:

  • Multiple packages on one page
  • Different colors for different pricing packages
  • The design of this page  (Consider using dummy branding)
  • Thinking about how you present your prices is actually very important and there are so many things that can be tested! 

Live Chat

With chat software, you actually have the opportunity to make customers feel like they aren’t being abandoned.

If your chat software has this function, then you should definitely test it with an A/B test!

 Not only that but having a live chat can increase conversions dramatically! 

There are two things that you want to test:

  • Which part of your funnel will the messenger exist?
  • Display name on messenger and avatar as well (this one is optional)

Video and images

This is probably the most important part of your website, but at the same time, it’s very easy to lose a user’s attention. 

You need to make sure that you present your product or service in the best way possible. 

You want to keep it simple and easy for users to understand through images and videos!

There are multiple things that could be tested here:

  • The order of images (1 vs 2 vs 3)
  • The use of text on the video or images (will people even read them?)
  • Use different product demonstrations like videos or photos  (yes, some people don’t like watching videos)
  • How many times should you display your product/service? What kind of pictures can I use?

Web copy

Very rarely do I see companies that actually test their web copy.  

Your website is your main source of converting users into customers, and it’s pretty important to get every word just right.

There are a few things you can test:

  • Headlines (make sure they are eye-catching)
  • Paragraphs and length of your text (readability and how easy it is to understand)
  • The language, tone of voice, audience expectation? 


All of these should be tested!  I’ve seen people use words that don’t even relate to what the product does…be careful with this one.

Top A/B testing tools: how to choose the best heatmap tool for your site?

Tools that allow you to create a heat map or evaluate heat maps include: 

Crazy Egg

FigPii

VWO

Optimizely

AB Tasty.

Conclusion

So to wrap all of this up, let’s go over briefly what we talked about in this blog.

We’ve answered the question of What’s A/B testing?

A/B testing is the experimentation that you do to determine which version of the same asset is better.

It’s randomized experimentation where a sample of visitors or users are shown a variant while another sample showed another.

It’s a competition where you bet A against B to see who wins based on your evaluation criteria, be it clicks, views, lead generation, etc.

We’ve also gone over tests that are closely related to A/B testing like 

Split testing

A/A testing

We’ve discussed how A/B Testing works, to have an overall idea not just about the concept but also about its fundamentals.

We’ve talked about why you should A/B test? How does it help your business not just improve its conversion rate, but also be more engaging with your audience.

We’ve expanded to talk about the A/B testing process, how does it work and what do you need to run a successful A/B testing campaign.

We’ve gone over some stats about how many A/B to run per week? or per month.

we’ve talked about A/B tests success rate and how come that a lot of your experiments fail, and is that a bad thing or not?

We’ve also discussed how long should you run an A/B Test?

And lastly, we’ve gone over which are the best elements to A/B Test on your website?

FigPii specializes in providing high-quality web analytics tools such as heatmaps, user polls, A/B testing, and session recordings so signup today for 14 days free trial period and see how we can grow your business together!

Heat maps by FigPii

Author