Solutions
Three sampling lines, one quality record
We field business audiences, consumer audiences and continuous API-driven studies from the same panel and the same screening pipeline. What changes between them is how respondents are profiled and priced. What does not change is the evidence attached to the file you receive.
Most sample problems are not fielding problems. They start earlier, when an audience gets described in words a provider cannot target and a buyer cannot check. “Senior IT decision-maker” is an adjective. Director level or above, 500+ employees, named on the purchase order for infrastructure spend is a specification. The second one can be verified; the first one can only be hoped for.
So each line below starts from the same question: which variables do we profile on entry, and which of them can you audit after delivery?
| Line | Profiled on entry | Typical incidence | Priced by |
|---|---|---|---|
| B2B audience | Occupation, seniority, company size, sector, purchase authority | Low, often under 5% | Incidence × length, per complete |
| Consumer | Age, gender, region, household, category behaviour | Moderate to high | Quota difficulty, per complete |
| API integration | Either of the above, plus your own screener state | Varies by study | Volume commitment |
What we will not do
We do not quote feasibility we cannot source. If your audience is a few thousand people worldwide and they are not in the panel, the honest answer is that you need recruited specialists rather than panel sample, and we will say so rather than take the project and miss the quota in week three.
We also do not route the same respondent through two studies in the same category inside a month. It raises throughput and it quietly destroys the data, because the second time someone sees a concept test they are no longer naive to it.