Test and learn
A monthly rhythm that turns ideas into tested evidence, and evidence into better briefs.
Why it matters
Without a way to test, measurement falls back on platform attribution, and agencies get rewarded for activity. A steady programme of tests gives you a better answer than the platform's report.
Monthly rhythm
- InsightsWe start with what the data shows this month.
- BacklogIdeas are ranked, and each one carries its strongest alternative explanation and how to test it.
- TestsEach test is registered with a hypothesis, set-up, test and comparison regions, tracker windows and watch-outs.
- Reading the resultWe say what's proven and what isn't.
- BriefsYour agencies get briefs they can act on, built on what worked.
- ReportIt ends with the decisions we need from you.
You leave with
A steady flow of tested ideas, and briefs built on what worked.
Sample · fictional data
Test register entry
| Hypothesis | Showing the brand in the first three seconds of the video lifts branded search. |
| Set-up | New edit in test regions, current edit in comparison regions, for six weeks |
| Alternative explanation | A competitor's promotion in the comparison regions |
| Watch-outs | A public holiday in week four |
| Result | Test regions ahead by 6%, within normal variation |
| Status | Not proven yet |
Decisions we need from you
- Run the test for four more weeks, or stop it?
- Approve a second region for the next test?
Where this comes up
When the board asks for the numbers, have the answer.
Tell us where it hurts: the budget conversation, the agency relationship, or numbers nobody trusts. We'll suggest where to start.
TALK TO NEIL →