A/B Test Card
A/B testing is an experiment in which two (or more) versions of a variable are shown to different random audiences at the same time to see which has the biggest impact. It is a powerful tool that can help us continuously improve.
You can summarize an A/B test in a simple card which includes:
- Product or service name you are testing
- Audience you are testing on (hint: be specific)
- Hypothesis (see below for some tips there)
- Conditions A & B (best to be visual and show the different conditions)
- A/B test results & learnings (to complete after the test is done)
It is important that you develop strong hypothesis. Above all, you need a hypothesis that is measurable and leads to action. There is lots to say about developing strong hypotheses (in fact there is books dedicated to business experiments). Below just one tip on different ways to write a hypothesis and its components:
- We believe that we can [outcome] by [what: action] [who: audience]
- We believe that [problem] will affect [who: audience] leading to [outcome]
- We believe that we can [outcome] by [what: action] [who: audience] because [why]
The A/B test card can be customized based on your needs. Here we see a version customized for a marketing application, which on top of the previous elements, also includes:
- Customer journey step (where is the customer in the journey)
- Business objective (what are you trying to achieve)
- Customer objective (what is the customer trying to achieve)
- Channel (where is the test taking place)
- Touchpoint (which specific touchpoint within the journey)
- Metrics (what will you measure)
- KPI (which metric will you use to decide between tests)
Go forth and A/B test…
Download the A/B Test Card tool
