How Recruitment Funnels Actually Work in Survey Sphere
From posting a study to a completed response: what happens at each stage, and where most of your applicants actually fall out.
A recruitment funnel is just the sequence a participant moves through between seeing your study and submitting a usable response. In Survey Sphere, that sequence has four stages, and almost every "why is my sample small" question traces back to one of them.
Stage 1: Visibility
When you publish a study, it becomes eligible to appear to participants who match your targeting — tier, category interest, and any eligibility criteria you've set. Visibility isn't guaranteed reach; it's eligibility to be shown. A study with narrow criteria and a small matching pool will have low visibility no matter how well everything else is configured.
Stage 2: Starts
A participant who sees your study and opens it becomes a "start." This is the number most teams watch first, but it's a weak signal on its own — a high start count with a low completion count usually means the study looked appealing from the outside but disappointed once opened, not that recruitment failed.
Stage 3: Screen-outs
Before a participant reaches your real questions, they pass through your eligibility screener. Anyone who doesn't match gets screened out here — this is expected and healthy up to a point. A screen-out rate that's unusually high for your category is worth investigating; it often means the criteria are stricter than the targeting that got them there, so people are starting who were never going to qualify.
Stage 4: Completions
Participants who pass screening and finish the study become completions — the number that actually matters for your data. This is what your recruitment funnel is optimizing for, and it's the only stage with a direct line to your study's payout budget.
Where funnels actually break
In practice, most underperforming studies break at Stage 1 (targeting too narrow for the pool available) or Stage 3 (screener criteria that don't match the audience the targeting actually reaches) — not at Stage 4. Before assuming a study "isn't attracting people," check whether it's attracting the *wrong* people first. The [Study Detail page](/sphere/studies) breaks these four numbers out individually so you can see exactly where a given study is losing participants, rather than only seeing the final completion count.