Formstack Blog

Do You Really Need a Form A/B Testing Strategy?

Conversion optimization is all about strategy, and A/B testing is no exception. If you try to tackle your A/B tests without defined goals and a solid plan, you will end up wasting time and money when you could be optimizing and making a profit.

Form A/B testing involves measuring different form elements against each other to see what converts better. And if you do it right, you can see some big results. According to the 2015 Form Conversion Report, just adding one word after “Submit” in your submit button copy can increase conversions up to 250%!

To ensure you get such stellar results, here are 7 steps you can take to build a successful A/B testing strategy for your online forms:

1) Analyze your situation.

You know the old saying “Look before you leap”? Well, take a look at your data before you leap into A/B testing. Determine your current conversion rates, measure them against the averages in your industry, and look for possible problem areas. Attacking your A/B testing goals with an optimized approach will ultimately heighten your chances of getting results that matter.

2) Remember that many factors can affect form conversions.

Don’t become too focused on one aspect of your form and forget all the other little bits that can make or break your conversion goals. Many elements can dramatically impact conversions, including design, copy, social media hacks like Facebook Tab Embed, and peak submission times. A/B testing all of these elements over time can greatly enhance your lead capture, but don’t forget that elements can be interconnected—so changing one can affect the others.

3) Test only one form element at a time.

Testing too many elements at once can muddle your results. Focus on one at a time so you can really narrow down on that element and see exactly how specific changes influence customer behavior.

Important form elements to test include:

4) Create a clear hypothesis.

A well-formulated hypothesis is the key to a good A/B test. Decide what change you want to make and think about what impact that change might have. This will ensure you can tackle your A/B test with a clear focus and an understanding of the results you are pursuing. And always remember, even if your hypothesis is disproven, you still learned something valuable about your test element.

 5) Distinguish between different user segments.

Don’t be too quick to disregard a test if it didn’t pan out. Some elements may test well with one segment of your customers, while other elements may fall flat. Make sure you know your user segments, and if possible, try to test those specific segments separately in your A/B tests.

At the very least, you should be analyzing your results to see if your testing element fared better in certain segments versus others. This way, you can find special strategic messaging opportunities for different users in your audience base.

 6) Evaluate your results.

A good benchmark for the length of an A/B test is when you get at least 100 people to both versions of your form. Only tests with a 99.9% statistical significance are valid, so you’ll want to check your test two weeks in to make sure this is so.

Once your test has run long enough, it’s time to look at your results. If the new variation of your form won, then congratulations! It’s time to make that version live and watch the leads pour in. However, if the new variation didn’t test well, don’t fret. You can take that knowledge and use it in future A/B tests.

7) Be aware of possible side effects.

When switching to a new version of your form after a successful A/B test, remember that there could be consequences. Tweaking one element for a higher conversion rate may lower the performance of another. It’s up to you to analyze the different elements of your form and decide which can be sacrificed for more submissions.


For more form A/B testing tips, click below to check out our free, interactive e-book “Testing Your Way to a High-Converting Form.” You’ll get A/B testing ideas, design tips, and proven ways to create copy that converts!

Happy testing!