User research panels fail in predictable ways. Nielsen Norman Group identifies the core problem: panels are treated as static databases instead of living systems. A list of participants that isn't continuously refreshed goes stale, and stale panels introduce bias that corrupts your research without announcing itself.
The piece is worth reading in full because it names specific failure modes, not just the obvious one. The static-database problem is the first of several described mechanisms by which a well-intentioned panel degrades over time, each producing a different kind of bad data.
If your team runs ongoing research using an internal panel, this is a maintenance manual, not a think piece. The practical stakes are high: accelerated studies and cost savings disappear the moment your participant pool stops representing your actual users.
[READ ORIGINAL →]