Low-incidence projects are where good research teams earn their reputation and bad sample does the most damage. If your study needs owners of a specific medical device, or procurement leads at mid-size manufacturers, every single qualified complete matters. And every fake one is expensive in more ways than CPI.

Key takeaways

  • Low IR plus a high incentive is exactly what fraudsters look for.
  • Start with profiled sources and targeted recruitment, not broad screening.
  • Tighten validation as incidence drops, never loosen it.
  • Agree realistic timelines and flexible quotas up front.

Why fraud loves low incidence

Screen-outs are expensive for fraudsters too, so they learn which answers qualify. A niche study with a generous incentive and a guessable screener attracts them quickly. The pattern is recognizable: a slow start, then a sudden burst of completes late at night from one region.

Sourcing strategies

Profiled sample. If a panel already knows who has type 2 diabetes, drives a pickup or approves ERP purchases, you screen far fewer people.

Targeted recruitment. For audiences no panel covers well, direct recruitment through relevant websites and communities works better than broad blasts. Our publisher-based sourcing exists partly for this.

Specialist partners. Some groups (physicians, certain trades) are best reached through specialists. A good provider says so rather than pretending.

Screening efficiently

Pre-qualify from profile data, then confirm with a short screener. Hide the qualifier in multi-select lists, as covered in screener design. Every question you cut from the screener improves feasibility.

Protect quality

For rare groups we add checks, not remove them: knowledge questions only a real member of the group would answer easily, consistency with profile data, and closer review of open ends. For professional audiences, see verifying B2B respondents.

Set expectations early

Low-IR studies take longer. Be honest with stakeholders about field time, agree which quotas can flex, and decide in advance what happens if feasibility comes in lower than planned. A provider that promises everything by Friday on a 2% IR study is telling you something about how they'll fill it.

FAQ

What counts as a low-incidence survey?

Generally anything below about 10% incidence among the people screened, and under 5% is considered very low. Feasibility and cost change sharply in that range.

How do you recruit hard-to-reach respondents?

Use profiled panels first, then targeted recruitment through relevant sites and specialist partners. Keep screeners short and add extra validation for the qualifying group.