How to Improve Landing Page Conversions: A/B Testing

Whew, you've finished your landing page design and now it's launched. You're done!

Well not quite. Don't miss an opportunity increasing conversions for your business. Consider A/B testing for your landing pages.

First, what is A/B testing for landing pages? Some people may know or have this idea of what it is. Others might not. Think of A/B testing like you’re a scientist experimenting. Experimenting between two, three or more pages to see what landing page converts the best.

Why should you do A/B testing? Simple. There will be differences in your target audience’s behavior when they land on your landing page. What works for one industry may not work for another. Same between different companies of one industry. One reason to start testing is if you have a high bounce rate. Or when paid traffic (PPC) isn’t buying. A/B testing is finding the best solution for you.

Okay, so what do you A/B test? Experiment with your landing pages. Test headlines, call-to-action(s), images or completely different designs. While testing is limitless you still want to consider on how you manage your a/b testing experiments.

How to do A/B Testing

Now you’re convinced that A/B testing is the key to higher conversions, more sales and the bottom line. One of the first things to consider is what page or goal you should test first. You may have different service offerings or products to sell. If you want to go more in depth you can test with different audiences if you need to as well. A good way to also look at it is by testing your Key Performance Indicators (KPI). KPI’s are measurable values. They show how effective you are achieving key business objectives. For instance, if you are a SaaS company you may have sign up rates as a KPI.

A key thing to note for testing and results is that you should have a good bit of traffic coming in already. The more traffic, the more data will make sense.

You need to make a list of tests

A/B testing takes setting up a strategy. Once you find out what page, goal or KPI to test then your next progression is creating a list of tests. An easy way to do this is to create a spreadsheet. A few examples of what to test are:

Headlines - It’s usually the first thing a web visitor sees. A compelling headline can make a huge impact. Using numbers in headlines has been considered most valuable for conversions. For instance, "30 Ways to Make BBQ Ribs". Also, a headline that makes sense and is clear to the reader is always attractive.

Copy for your Call to Action (CTA) - Like headlines the action you want users to take is important. You’ll have different reactions when your call-to-action says “Download our free eBook today!” or “Our eBook is a must read now”. Being direct with a command will let your visitors know what you want them to do. The CTA with the exclamation point provokes enthusiasm which can make your visitors excited to take action.

Call to Action Placement & Styles - Place your call-to-action in different places. It will produce different results. Along with how it looks like. You may want one test using a button and the other as a text link. Or your button tested in different colors.

Images and Graphics - Don’t overlook the importance of what visual elements look like. Use images that signal an emotion. Even graphics have the same effect.

Landing Page Copy - A/B test short and long copy. Remember what may work for one audience might not work for another. For example, an audience looking for mortgage rates might not need to be bombarded with long copy. While an audience comparing software tools may want a lot of information.

Pricing - While you may not be able to change your prices. You can definitely test how you present it. Telling people about free shipping or a percentage off is different even if it’s the same costs to you.

Layout - Test changing layouts of sections or elements. Layout 1 sees the header, image banner with a CTA, testimonials, trust factors, sale copy and footer. In that order from top to bottom. Layout 2 sees the header, trust factors, image banner with a CTA, sale copy, testimonials and footer.

Remember you’re testing one thing at a time. If it’s the headline then let the tests be about the headline. Landing page layout tests can have completely different designs between the variations. Which is focused on seeing what design performs better.

Have tests in order of priority

Now you have your list of tests. It’s time to put your A/B testing items in priority. The key reason? Test things that will have the biggest impact first.

If you have a bunch of tests lined up it’s not going to do any good if you are confused on what to test first. By creating a process of what to test first and asking yourself why are these test placed in this order.

Going back to KPI’s is a good place to start when prioritizing. The pages that bring in sales should be the first. Go ahead and test your checkout or sales pages as a priority. Improvements on these pages can lead to the biggest revenue changes.

Another priority tactic is to focus on the bigger wins. Changing your message on your CTA message can create a higher conversion rate. The biggest wins don't have to be something drastic. In fact, it’s good to have a test that doesn’t have much effort to change but returns a high reward. Such as the CTA message for filling out your form.

Like the list of A/B tests, you need to have a process and framework on prioritization. One way to do this with each test is to figure out a few things such as:

  • Is there a lot of creative work to do?
  • How complex or difficult it is to create the test?
  • What is the impact of the test?

Know the design process of your A/B test

You’ve completed your list and everything's organized. Take a similar approach to A/B testing like any web design and development process. How is all this going to get completed? Who’s creating or making modifications to test pages?

We highlight an important aspect when choosing your option below:

Freelancers/Contractors - Do they have experience in optimizing landing page conversion? Ask upfront if they can design & develop, or design. Learn about their schedule as well to know if they can work on your testing.

In-house team - Are you a company with an in-house team? Know if they already have experience with conversion rate optimization design & development. If not, then you may want to consider other outside sources to with your team or on the project itself.

Conversion Rate Optimization Agency - Hiring a team allows complete service and expertise. Test ideas, design variants, development, running and analyzing results are off your plate. Discuss your goals through consultation and recurring meetings.

Yourself - If you don’t have the liberty of any above your next options would be to use landing page tools. Ones that provide A/B testing features. A/B testing tools like Unbounce. If you already have a website. For instance, a WordPress website there are plugins that will help assist with A/B testing.

During the A/B testing make sure to run both tests simultaneously. Give the tests time to collect data. You can base the time on the amount of traffic you’re getting. At least 10,000 visitors is a good starting point to give you a sample size. If you get 10,000 visitors in a day, a week or month will also determine how fast you can conduct tests.

Alright Professor now it’s time to analyze your A/B testing results

Depending on what route you’ve taken with your design process. You’re either using an A/B testing tool that provides statistics or whom you’re working with will.

Focus on your goals and KPI’s. Paying attention to your primary test goal you’ll be able to see the results and determine what are the next steps. While you run these tests make sure that the results are significant enough to demand a change. Let’s say you’re performing a test on the value of CTA button colors. Test A button has a click rate of 5% while test B button is 20%. You’ll know what that will mean to you to make a uniformed decision.

The next step is to move onto your next test and process again!

Need a team to effectively A/B test your landing pages? Get in touch with us.