Split Test

Evidence Strength*:

3


Time/Cost: 

2


skills Required: 

User Experience, Product Design, Development, Data Analysis

Compare two versions of something. A versus B to determine which performs better.
Quick how-to guide:

Split test is perfect for testing different versions of value propositions, prices or features.

  1. Choose customer behaviour you want to improve.
  2. Create Control A and Variant B.
  3. Choose your customer sample size.
  4. Run test: randomly drive 50% of traffic to each.
  5. Analyse results: Did you confidently see a result? If not, run a different test with new Variant B.
need a team to help?

*Evidence Strength is the expected value of evidence the experiment will obtain. Measure this against the time/cost estimation to decide which experiments to undertake. Rating is between 1-5 and this is a rough estimate. Your business situation and market is unique. For more help or advice get in touch.