There are plenty of reasons why tests become inaccurate:

• implementation of the hypothesis didn’t match the test hypothesis
• change was too small and potentially not even seen on the page (below the fold)
• test duration and/or sample size quality (SRM or MDE issues)
• etc.

By combining test results (aka meta-analysis) with similar behavioral hypotheses and psychological directions, you can see the trends, patterns, and validity in your testing program. Watch over Ryan’s shoulder as he shows swipeable Airtable examples of how he documents tests and explains simple meta analysis to shows the learning and insights that all test programs should have.Ryan will also walk through the Hierarchy of Evidence (adapted for CRO) to show what decisions have the least risk across a CRO program.

Example graphic: https://rednavel.link/CyG9Jvd8