How to create an A/B test for forms in Klaviyo
You may be wondering: what factors drive the best conversion rates for your forms? Is it specific wording, the call-to action button, or even the colors of the form?
Create an A/B test in Klaviyo to test different variations of your sign-up form and identify what works best for your audience. You can create multiple versions of your form to determine what design, messaging, or form fields are most effective. Avoid making assumptions and instead make decisions based on actual data.
Best practices for A/B testing forms
Before setting up an A/B test, clearly define your goal. Determine the specific elements of the form you want to test and the desired outcome. This could be: testing different images on my form to increase form submission rates.
Let's run through key A/B testing strategies and learn how the coffee company, Beantown, uses A/B tests to make data-driven decisions.
Test 1 element at a time
First, focus on identifying 1 core element of your form to test. This makes it easier for you to attribute positive or negative results to a specific part of the form.
The element that Beantown will test must align with their goal to improve form submit rate. They want to know if the design of their form is causing low submission rates. For example, could the type of images, font size, or copy of the form be contributing to the lower metrics?
To start, they will test the form's image to see if different graphics have a substantial impact on form submissions.
Create 2-4 distinct variations
Next, create multiple form variations to test.
Beantown chooses to test 2 form variations, and they label each with a specific description based on the element being tested:
- Variation A: Images with products
- Variation B: Images with customers in store
Keep in mind, Beantown can add more variations, however the test will need to run for a longer period of time.
Determine traffic size and duration
Next, decide what percentage of people will see each variation, as well as how long you want your test to run.
By default, Beantown will use an automatic weight, so each form variation will appear to an equal amount of viewers (i.e., 50/50 split). This leads to more accurate and reliable results. That said, Beantown can also manually adjust the percentages, as long as it equals to 100%.
The last step is to determine the duration of their test to ensure it runs long enough to capture sufficient data for analysis.
How to choose a winner?
Beantown chooses the automatic option to end their test. The test will end when 1 variation outperforms the other at a statistically significant level or when it reaches a particular end date. If they choose to edit their test and end their test manually, Klaviyo will prompt them to choose both the winner and the date that the test ends.
Implement changes based on winning variation
Monitor and measure relevant metrics in the form's Overview page to track the overall performance of their form during and after the test. Additionally, take the next steps to get accurate data.
1. Track test results
Beantown makes sure to review the A/B Test results page to dive into the analytics of their test. Here, they can see which variation outperforms the other.
2. Draw conclusions
Once their test has concluded, they will analyze the results and draw conclusions based on the data. They must identify any significant differences between variations and determine which resonates more with their audience.
3. Iterate and continue testing
Beantown will use the insights gained from their A/B test to make informed decisions and continuously improve their forms. If Beantown determines that visitors prefer images of real customers, they can adjust future forms to use those images. If the results are inconclusive, Beantown can make new variations or test a completely new element. They can create as many tests as they'd like to further improve their form's effectiveness.
Looking for additional guidance? Bookmark how to review A/B test results.
Run an optimization test
Find the best display time for forms on your website, so you can effectively grow your subscriber list. Similar to an A/B test, an optimization test identifies areas that are working well or need improvement. Klaviyo runs a series of AI-generated experiments to determine the most effective display timing settings that increase a form’s submission rate. As a result, Klaviyo will automate this test to save you time and deliver more accurate results.
Note: You must have at least 400,000 profiles (note that this is total profile count) to qualify for optimization tests.
For step-by-step instructions, bookmark how to optimize a sign-up form's display settings.