Reference table
Stopping rules by study type
A planning table for stopping rules in product and UX research. For surveys, concept tests, qualitative discovery and continuous discovery programmes it lists what to fix before fielding, when to stop early because the answer is clear, and when to stop because the study cannot reach a useful answer. Put the rule in the research brief before the first response arrives, so any interim look is a planned one.
Where this appears
What's in this reference table
| Study type | Pre-specify this | Efficacy stop | Futility stop |
|---|---|---|---|
| Survey with a primary metric | Target n, look schedule, boundary family | Result crosses the early boundary | Confidence interval already excludes the effect you would act on |
| Concept test across variants | Primary variant comparison, look schedule | One variant wins decisively at the interim look | No variant separates from control by the midpoint |
| Qualitative discovery | Codebook, new-code rule, floor and ceiling | Three consecutive interviews add no new code | Ceiling reached. Write up what you have |
| Continuous discovery programme | Review cadence, the decision the research feeds | The decision it informs has been made | The decision was cancelled or made without it |
Use this image
Free to use in your docs, decks and posts under CC BY 4.0, as long as you credit Koji with a link to this page. To embed it, paste this into your page:
html
<a href="https://www.koji.so/visuals/stopping-rules-by-study-type-table"><img src="https://www.koji.so/visuals/stopping-rules-by-study-type-table/stopping-rules-by-study-type-table.png" alt="Table of what to pre-specify for four study types, from surveys to continuous discovery, and when each should stop early for success or for futility." width="1200" height="766" /></a>
<p>Source: <a href="https://www.koji.so/visuals/stopping-rules-by-study-type-table">Stopping rules by study type</a>, Koji</p>
