Use experiment canvas in CRO

Contributors

@paul-boag


Business Benefits

Formulate a clear hypothesis that is ready to be tested and run effective and efficient experiments in a common language.


Form a hypothesis you wish to test based on available facts.

Avoid forming hypotheses based on intuition, but instead, use the data available to you to help identify what you should be focusing on. For example, analytics might indicate a certain page is under-performing and by watching session recordings of that page you might be able to establish a hypothesis of why this might be the case. By sticking to the facts you make it more likely that your marketing efforts will be effective.

Always try to base new marketing ideas you want to test on the most trustworthy facts. It’s a closer look at the validity and reliability of the research methods. How valid the facts are is always subject to discussion, and validity should be based on whether the research answers the questions you’re after.

Clearly define your desired behavior and how you are capturing it in your data.

How are you going to demonstrate that your hypothesis is correct, what action does the user need to take, what data will inform you that the desired behavior occurred in your experiment, and how can you tell it apart from regular customer behavior?

For example, what data will inform you that the product the customer just purchased, is actually an additional product added to their shopping cart in response to your hypothesis, and not just a purchase they would have made anyway?

Choose a trigger or CTA for your desired behavior.

Your trigger is the step you want users to take to prove your hypothesis true or false. For example, if you believe that a different page title would increase the number of sign-ups to your newsletter, then the trigger would be the newsletter sign-up form.

When picking a trigger, try to pick one that is closely linked to the problem. For example, if you hypothesize that the page title negatively impacts newsletter signups, make sure the newsletter signup is closely associated with the page title. Failing to do this will mean fewer users reach the signup, meaning it will take longer to get statistically accurate results.

Choose a highly motivated and receptive audience.

Focus on forming experiments where users are most likely to take action. That will have the biggest impact on conversion, but also make testing easier. At what stage in the customer journey is your audience most receptive or motivated to take action? For example, in the famous Angry Birds game, users are typically most receptive and motivated to take action when they get stuck on a level.

When your hypothesis requires a new call to action be sure to pick an appropriate touch point at the right point in the user journey.

It is important to introduce new touchpoints at the most appropriate moment in the user journey. For example, you may hypothesize that a pop-up overlay may increase newsletter signups. That may be correct. However, if you display the overlay too soon, users may not be ready to act, and so you will conclude (incorrectly) that an overlay is not the answer. It may have worked at a different time in the journey.

Complete the hypothesis, “If I [trigger] [audience] via [touchpoint], then [desired behavior] appears in [data].”

This, along with your filled in Experiment Canvas, can be used as the basis for your next experiment and just needs to be translated to an actual piece of content or feature for your website.