Have you heard about A/B testing? It's a method that shows how small changes can have a big impact on your product's success.

For example, will a more eye-catching call-to-action (CTA) perform better than the current one? Or maybe changing the button color, font, form placement, or language used will increase conversions?

In A/B testing, you divide users into two groups: group A and group B.

  1. Group A is the control, where 70% of users experience the original version.
  2. Group B is the treatment, where the other 30% of users see the developed alteration.

How do you determine the best percentage for an A/B test? It depends on the sample size and objective. Some suggestions include:

  1. 50/50, 70/30, or 80/20 are ideal for testing two versions with equal importance and large samples.
  2. 70/30 or 80/20 is useful for testing more significant changes with lower risk.
  3. 90/10 or 95/5 is recommended for risky tests or small changes, minimizing negative impact.

Here are some examples that illustrate the various applications of A/B testing in different business areas:

  1. A/B tests for new features: A social media company tests the addition of new reactions to their posts to see if it increases user interaction.
  2. A/B tests for interface changes: An e-commerce website tests different designs for the "Buy Now" button to see which one results in more purchases.
  3. A/B tests for communication and design hypotheses: A music streaming platform tests different email titles to see which one attracts more subscribers.
  4. A/B tests for page optimization: A hotel reservation website tests different versions of the checkout page to increase the booking completion rate.

It's worth noting that A/B testing is not recommended for creating completely new experiences, such as a product that completely changes its features or user experience. In such cases, it's necessary to allow users to get used to the change, especially recurring users.

Also, for reliable results, it's essential to have a statistically significant sample and an appropriate duration to collect relevant data.

Now that you know what A/B testing is, why not try it out and see how small changes can generate a big impact on your product's results?


Jordan Chicote

Passionate Product Owner/Manager driven by data and results. I deeply value collaboration with teams and understanding customer needs. I have extensive experience working on diverse projects in different segments. Since 2016, I've been leading the development of innovative solutions in the product area. Skilled in strategy, project management, data analysis, and leadership.

Value Your Time: 11 Tips for More Efficient Meetings


Tips for Organizing Documentation in Agile Projects


Synergistic Digital Transformation: Potentiating Results with Systems Thinking