All frameworks and cheat sheets

Reference table

Quantitative research sample sizes by method

Minimum sample sizes for surveys, A/B tests, card sorts, tree tests, SUS, SEQ, MaxDiff and NPS

Table of minimum sample sizes for eight quantitative research methods, from 385 survey responses to 15 or more participants for SEQ.
Check your plan against this before launch. Across 80 users, a 5% lift is noise.

A cheat sheet of recommended minimum sample sizes for eight quantitative user research methods. Surveys need about 385 responses for a ±5% margin at 95% confidence, card sorts 30 or more and tree tests 50 or more, following Nielsen Norman Group. SUS needs 30 or more, task-level SEQ 15 or more, MaxDiff and conjoint 200 or more, and NPS, CSAT and CES hundreds to thousands per cohort. Use it to check a study plan before you launch.

Where this appears

What's in this reference table

MethodRecommended minimumNotes
Survey (±5% margin, 95% confidence)385For populations over 10,000
A/B test (small effect)1,000s per variantDriven by baseline rate and minimum detectable effect
Card sort30+Per Nielsen Norman Group
Tree test50+Per Nielsen Norman Group
SUS30+For a stable composite score
SEQ at task level15+Read alongside completion rate
MaxDiff or conjoint200+Higher when ranking many items
NPS, CSAT or CES100s to 1,000sPer cohort

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/quantitative-research-sample-size-by-method"><img src="https://www.koji.so/visuals/quantitative-research-sample-size-by-method/quantitative-research-sample-size-by-method.png" alt="Table of minimum sample sizes for eight quantitative research methods, from 385 survey responses to 15 or more participants for SEQ." width="1200" height="885" /></a>
<p>Source: <a href="https://www.koji.so/visuals/quantitative-research-sample-size-by-method">Quantitative research sample sizes by method</a>, Koji</p>