You are recruiting for research on a public-facing benefits checker. The delivery manager offers to fill the sessions from your own support team, who use the tool every day, and suggests screening by asking each participant whether they have a disability. What does the Service Manual say about the two proposals?

The principle — two recruitment decisions, and both of them decide what the research is capable of showing before a single session runs.

Why the key is correct — on staff, the manual is direct: Avoid using your own staff for research into the public facing parts of your service. The reason it gives is the one the delivery manager has mistaken for a qualification, since Their knowledge and experience can mean they will use the service in a very different way. A support team that uses the checker daily has internalised every workaround, so it can tell you what is annoying and cannot tell you what is confusing. The requirement stands separately too: For your research to be effective, your participants must be actual or likely users of your service. and a benefits claimant is not a support agent. On screening, the manual warns that Some people with access needs do not identify as having a disability and it is important to be aware of this in the screening process. A question that asks people to identify as disabled therefore filters on self-description rather than on need, and it misses the person who has a tremor, or reads slowly, or is using the service on a phone in a noisy room. Screening on what people need in order to take part reaches all of them.

Why the others are wrong — treating daily users of an internal tool as ideal participants for a public service inverts the manual's reasoning, since familiarity is the disqualification rather than the credential. Declaring either group acceptable so long as screening is consistent confuses asking the same question of everyone with asking the right question at all, and it ignores the requirement about who participants must be. And accepting the disability question as adequate cover keeps the wrong instrument, which only ever reaches people who use that word about themselves.

Remember this — recruit actual users of this service, and screen on what someone needs to take part rather than on what they call themselves.

Sources — the UK Government Digital Service Manual on finding research participants.