A/B testing is a fundamental process that allows you to test areas of various elements of your website to see which techniques or designs drive the best results for your company.
But to conduct a test that allows you to gather the data you need to make revenue-driving optimizations, you’ll need to choose the best A/B test sample size.
Luckily for you, we’re going over how to calculate the sample size for your A/B testing in this blog post, so just keep reading!
And don’t forget to sign up for our free newsletter while you’re here to join over 200,000 other marketers who get expert marketing advice and updates sent straight to their inboxes!
Don’t miss our Marketing Manager Insider emails!
Join 200,000 smart marketers and get the month’s hottest marketing news and insights delivered straight to your inbox!
Enter your email below:
“*” indicates required fields
(Don’t worry, we’ll never share your information!)
Understanding A/B test sample sizes and timeframes
Perfecting your A/B test plan is no easy feat. So, before we dive into how to calculate your A/B test sample size, we’ll break down how the testing process works.
When you conduct an A/B test, you’re essentially testing an area of your website or marketing strategy.
For example, let’s say you want to test two versions of your email subject line to see which one results in the highest open rate.
In that case, you’ll show the first subject line to a portion of your email list labeled “A.” Then, you’ll send the second subject line to another portion of your email list labeled “B.”
After the test, you’ll send the subject line with the highest open rate to the rest of your subscriber list.
Once you’ve decided what you want to test, your next step is determining your sample size and timeframe.
What is an A/B testing sample size?
A/B testing sample size is the number of people that will see the element you want to test. From the subject line example above, your sample size is the number of email subscribers you will send the two variations of your subject line to.
Sample size can vary depending on the element you want to test. For example, you might have a larger sample size when testing elements of your website, like call to action (CTA) graphics or headline copy.
On the other hand, if you’re testing your email subject line, you might want your sample size to be smaller so that the subject line with the highest open rate goes out to the majority of your subscribers.
What is an A/B testing timeframe?
A/B testing timeframe is the amount of time your test will run. Like your sample size, the timeframe can also vary depending on the element you want to test.
For example, you might test a new color for your CTA button for a month or longer. On the other hand, you would likely only run your subject line test for an hour or two before sending the winning one out to the rest of your subscribers.
How to calculate A/B test sample size
As we mentioned above, one sample size or timeframe won’t suit every test you run. So, let’s dive into how you can actually determine the best sample size and timeframe.
Here’s how to calculate A/B test sample size:
- Consider whether you have a big enough contact list for a sample
- Use an A/B test sample size calculator
- Enter your conversion rate and minimal detectable effect into the calculator
- View your results
- Calculate the percentage of your sample size if needed
1. Consider whether you have a big enough contact list for a sample
The key to performing a successful A/B test is to get statistically significant results that enable you to gather in-depth insights.
In other words, if you conduct a test on a very small number of website visitors or email subscribers, you won’t be able to gather enough meaningful data to determine which version of the test was most successful.
It’s also essential to test the smallest portion of your total email list or web visitors to get significant results so that you can implement the winning test to the majority of your audience.
So, to A/B test a sample of your list or website visitors that will give you meaningful results, you’ll need a decently sized email list or amount of web traffic.
We recommend having at least 1000 audience members to conduct your A/B test. For anything smaller than that, your sample size would need to be the majority of your total audience to give you enough meaningful results.
And if that’s the case, the remaining number of audience members that will see the winning element from your test will be so small that you might as well have implemented the new change across your website for your entire audience to see what happens.
2. Use an A/B test sample size calculator
Your next step in calculating the sample size of your A/B test is to find an A/B test sample size calculator. Instead of learning an A/B test sample size formula, it’s a lot easier to use a free calculator to do the work for you.
There are a plethora of handy calculators out there that make determining your sample size a breeze, like this one from Optimizely.
Some calculators will also help you determine the perfect timeframe for your test. Once you’ve found a calculator you like, it’s time to enter some details to get your sample size.
3. Enter your conversion rate and minimum detectable effect into the calculator
If you use the A/B test sample size calculator from Optimizely, you’ll need to enter two details before you can learn the best sample size for your test:
- Baseline conversion rate: The baseline conversion rate is the expected conversion rate of your control group. Your control group is the number of people who will view the current element that’s on your site or marketing message.
- Minimum detectable effect: The minimum detectable effect is the minimum relative change in the conversion rate you would like to be able to detect from the new element you are testing.
4. View your results
The next step in how to calculate A/B test sample size is to view your results.
After you’ve entered all the necessary elements in your chosen calculator, it will reveal the optimal sample size for your test.
This is the number of total audience members you will want to test your new element on to get meaningful results that enable you to optimize your strategies for the future.
5. Calculate the percentage of your sample size if needed
Depending on your A/B test and the calculator you use, you might need to calculate the percentage of your A/B testing sample size.
For example, if you want to run an email subject line test, you might need to choose the percentage of your total email subscribers you want to run the test on.
To calculate this number, use this percentage A/B test sample size formula:
Sample size / 1000 = Sample Size Percentage
Ta-da! You’ve now got your sample size and can start your A/B test to begin optimizing your website and marketing strategies for more leads and conversions!
How to choose the best timeframe for your A/B test
Now that you know how to determine the best sample size for your test, how can you find the perfect timeframe for how long your test should run?
A/B testing timeframes can vary depending on the element you want to test as well as your own business goals.
For example, suppose you want to implement a new CTA button on your website by the beginning of the new year. In that case, you’ll probably want to finish your A/B test by October or November so you can implement the winning version on your website on time.
If you’re sending an email A/B test, it’s usually a great idea to look at your past email data to find the best timeframe for your test. Look at your data to determine what time your clicks and opens begin to drop off.
Then, ensure you run your test and send the winning email version to your entire subscriber list a few hours before that time so you can maximize your clicks and opens.
Time to Level Up Your Sales
Our long list of services helps you grow every aspect of your business with marketing strategies that are proven to increase bottom-line metrics like revenue and conversions.
In the past 5 years, we’ve managed more than 12.9 MILLION transactions across our client base.
Need help determining your A/B test sample size?
Running the perfect A/B test that reveals the data you need to improve your campaigns for better results is no easy feat.
If you’re finding it tricky to wrap your head around choosing the best A/B testing sample size or simply don’t have the time to run and analyze tests, the digital marketing specialists at WebFX can help.
Save valuable time and implement data-backed marketing strategies with a team that has over 1.6 million combined hours of experience.
Contact us online or call us at 888-601-5359 to learn more about how our team can help your company craft marketing strategies that drive revenue!
Join our mission to provide industry-leading digital marketing services to businesses around the globe - all while building your personal knowledge and growing as an individual.