A/B Testing Landing Pages in 5 Simple Steps
12th December 2019
A/B Test Your Landing Pages With These 5 Simple Steps
You’ve created a great landing page for your digital advertising campaign and have started seeing results! But is this the best you can do?
With A/B testing landing pages, it’s easy to test different copy (headlines, a call to action), personalization and voice, value proposition, button colour, page flow, media, and keywords to see which leads to more conversions. The key, however, is to test one variable at a time! Be strategic, test one thing at a time and see what makes a difference.
Ready to start – A/B testing in 5 simple steps:
1. Choose a tool
You can run an A/B test without the help of a tool, but personally, we recommend using Google Optimize. It’s a free, easy-to-use tool that requires only a small snippet of code to be added to your site to be able to run A/B tests. The program allows you to change text, images, form placement, and even HTML and then select the percentage of visitors you want to see the B version of the landing page.
If you decide not to use a tool, check out HubSpot’s testing kit with templates, calculators and a best practices guide.
2. Choose what do you want to test
Knowing what you want your visitors to do when they land on your page will help you choose what to A/B test.
More than likely the metric you want to increase on your landing page is conversions, but how are you defining a conversion. Is it a form fill, an email submission, etc.
As we mention before, you want to conduct only one test at a time. That means only testing one hypothesis, one variable at a time will show definitively what change made a difference.
3. Define the “A” Version
You will need to define the original or “A” version of the test. This, in scientific terms, is called your “control”.
For example, if you are hypothesizing that by moving the lead generation form on your landing page above the fold will generate more leads:
A Version: Control (will be the unchanged, unaltered, original version of your landing page)
B Version: A copy of your Control (A Version) of your landing page, but with the lead generation form above the fold.
4. Create the “B” Version
Now that you have a hypothesis and have chosen what variable to test, you can build your “B” version of the landing page and set up how long you will run the test.
If you are using a tool, like Google Optimize, it will automatically suggest a timeframe to test, typically 2-3 weeks.
Now let your test run and wait for the results!
5. Analyze the results
Did moving the lead generation form to above the fold generate more leads? If yes – then your “B” Version won!
Google Optimize will automatically show you which test won, or if it was inconclusive.
Now what? …
6. Test Again
Yes – we know there were 5 steps, but we want to make sure that you continue to test and refine your landing page so you can generate more results!
One last thing – to make sure you start strong and create great landing pages from the start check out our blog: Landing pages that convert.