A/B Testing: Your Guide

Conversions are likely the main reason your firm created a website. Getting a visitor to submit their details and transition into a lead is the goal. Work to boost conversions will undoubtedly be a focus for your firm.

But where do you start?

You may have theories about what you could change on your site to secure more conversions. But there could be consequences if you implement the change and find that the results are not what you were hoping for. And no organisation wants to see conversion rates fall.

If you’re torn between two – or more – concepts for a page on your site, A/B testing can help you identify the more successful option.

Here’s your firm’s guide to A/B testing.

What is A/B testing?

A/B testing is sometimes known as split testing. It’s a tactic used to compare conversion rates for two versions of a page on your website. You can use it to test design elements or page content. It’s commonly used to test conversion rates of calls to action (CTAs).

When you conduct an A/B test, you create a new version of an existing page. You then send half of that page’s visitors to the control page – A – and the other half will visit the new variation page – B. You then monitor the conversion rates to identify the version that performs best.

Once you know which version of the page your visitors prefer, you can use that information to attempt to improve conversion across your site by implementing the successful elements from your test.

It’s important to leave your test running for long enough. If you see results skewing one way and conclude the test too quickly, you may find that it was just a fluke and that the results weren’t truly reflective of what visitors would have done if the test had continued running.

You can test your website, email campaigns or paid advertising. Each platform will have its own way to conduct a test.

What to test

Your test can be as straightforward as changing the colour of your CTA button to make it more prominent or it could involve entirely rewriting a page to see if visitors connect more with the new content. It could be changing the layout of certain elements on a page or it could be implementing a new feature, like an email sign-up field.

You can test anything you think might have an impact on conversion rates. But consider your options carefully. It could be a waste of time to proceed with a test without a data-driven reason for it.

Evaluating your traffic could give you a shortlist of pages to target, while other analytics could help you narrow down the elements you want to change. You could look at heatmaps to see how users interact with your page and establish what might perform better.

A/B testing is essentially just data collection. First, you analyse your data to identify the right pages and elements to target. Second, you gather the data from your test. Third, you find out if the data you’ve gathered is applicable to other pages.

How to conduct an A/B test

It’s important to remember to test just one element at a time. Although you may come across a number of features you want to change, it’s not a good idea to do so simultaneously. Isolating and testing one specific change will allow you to establish if that individual variation works or not.

If you’re testing multiples together, it will be far more challenging to work out which element was responsible for any uplift in conversions. On occasion, you may want to test more than one variation. This is called multivariate testing and is typically only recommended for sites with high amounts of daily traffic.

Once you have your test results, you’ll need to establish whether they are statistically significant enough to implement any long-term changes. If the results aren’t convincing enough, it may not be worth the money and time needed to make these alterations.

A/B testing tools

You’ll need to rely on certain tools to conduct an effective A/B test. These can be free or paid – your budget doesn’t have to prevent you experimenting.

If you’re testing elements of an email – such as a subject line – your email platform will allow you to send one version to half your subscriber list and another to the other half. Meanwhile, if you’re advertising on social media like Facebook, you can set up A/B testing through the platform’s ad manager.

When testing your site, you can look at several free data sources. These could include:

Your Analytics can help you identify the pages you want to experiment with, as well as what KPIs you want to achieve. This might be reducing bounce or exit rate or boosting numbers of people filling out a certain form, which should be an established goal.

Google’s Marketing Platform comes with Google Optimize, which is the tool that lets you create your A/B test. It’s a simple and free tool that can give you the data you need to make future decisions. Check out Google’s guide to using Optimize to ensure you’re getting the most out of it.

Hotjar offers a free version, which lets you create three heatmaps and view up to 300 recordings of visitor interactions. Look at what people are clicking on, ignoring or how far they scroll. This could be enough to establish what elements of a page you could work to improve.

If you’re looking at paid tools, you could consider:

AB Tasty is an all-encompassing tool that lets you supercharge your A/B testing. It helps you build test pages, segment your audiences and analyse the results to use to your advantage.

Helping you make data-driven decisions, Optimizely is a well-known tool you can use to test variations and learn what visitors to your site want.

Unbounce lets you build and test different versions of landing pages to find the most effective one. This can ultimately help to reduce the amount you spend on PPC.

The full version of VWO is a paid program, but it offers some hugely useful tools for free. This includes a significance calculator, duration calculator and sample size calculator.

Attempting to improve your site's conversion rates through A/B testing is always a good idea.

And if your firm is looking to secure high quality leads, talk to the team at First4Lawyers. We're here to help you generate the work you want.


It seems you are using an outdated browser.

This will impair your browsing experience around the web. Please visit one of the links below to update to a modern browser then re-open the site with the new browser.

Thank you

Download our White Paper

Download our White Paper

Can't find what you are looking for?

We are open as normal during the Coronavirus lockdown and are able to help with all your legal needs.

Call us free of charge

0800 567 7866

Request a Callback

Continue browsing