Product & Project Delivery
Recruit Product Research Participants Who Match the Decision
Create a recruitment brief and screener around relevant behavior, then compare a completed fictional selection register with the target. Keep eligibility, arrangements and actual participation separate.
Recruit participants whose experience can answer the product decision you need to make. Define the relevant task, situation and behavior before choosing a channel or asking colleagues for names. Then check who actually meets those criteria, who can participate in the planned format, and which perspectives your current list still lacks. Keep sample fit visible alongside scheduling progress.
Start with a recruitment brief and a short screener, then maintain a gap register that distinguishes eligibility, logistics and completed participation. This guide from LandOffer.ai supplies both artifacts through one fictional product-research case. Every person, response, selection and confirmation below is authored teaching material. No one was contacted, screened, booked, recorded or studied, and no product account or private customer data was accessed.

Identify the experience your decision requires
ProofLane is an invented online print-proofing service. Its team wants to investigate how client coordinators understand the status of a replacement file before final print approval. The immediate decision is which status questions need further investigation before designing a change. It is not whether clients like a proposed redesign, how many customers would buy it, or whether a new interface meets an acceptance threshold.
Researcher Mina needs people who have personally attempted to replace a proof file in an online proofing service during the previous six weeks. Current ProofLane users and people using another comparable service may both offer relevant experience, but their service context must stay visible. An approver who only signs off files uploaded by someone else does a different task.
The GOV.UK Service Manual's participant guidance recommends involving actual or likely users and reviewing recruitment criteria against the research questions. It also warns that channels and scheduling can shape who is included. These principles help define a relevant pool; they do not make a small volunteer sample representative of every customer.
Mina separates necessary experience from convenient preferences. Personal replacement experience is essential for this round. A particular job title, enthusiastic feedback about ProofLane or fluent design vocabulary is not. Other questions might require approvers or administrators, but adding them now would change what this round can address.
Complete the brief before writing the screener
The fictional brief names a first-round target of six participants: three who completed the replacement without live assistance, two who stopped before completing it, and one who completed it with live help. For classification, Mina uses each person's most recent qualifying attempt. These three paths are mutually exclusive in this register; communication or access arrangements are separate fields.
Six is an illustrative planning choice for a focused qualitative round, not a universal sample-size rule or a statistical confidence claim. The team would revisit the plan if it discovered a materially different user group or could not answer the intended question. Merely filling six places would not establish saturation or population coverage.
Mina proposes 45-minute remote sessions during November 9–13, 2026, using authored example proofs rather than participants' customer files. Those are future planning assumptions, not approved sessions. A different format remains possible if needed. The brief asks the recruitment owner to confirm available times, participation arrangements and the organization's incentive process before issuing specific invitations. No payment amount or recording permission is assumed.
The GOV.UK recruitment-brief guidance calls for written instructions covering the intended participants and session arrangements, and reviewing an agency's screener against the brief. That last check matters even with an experienced recruiter: a standard filter for confident software users could remove people whose difficulties are important to this study.
The brief explicitly excludes the ProofLane design team from this round because of prior knowledge of the proposed work. It does not exclude every software professional, every colleague or anyone needing assistance. Eligibility follows the research question and relevant knowledge, rather than a generic judgment about who is a “good participant.”
Use a screener that finds behavior without teaching the answer
Nielsen Norman Group's screening guidance recommends relevant past behavior over predictions of future use, avoiding leading questions, and limiting screening to criteria that matter. A question such as “Do you find replacement-file status confusing?” would select people who agree with the team's suspected problem and reveal what answer is desired.
Here is Mina's completed screener specification. The routing instructions are internal research notes, not an eligibility answer key shown to respondents. Each question includes an honest “not sure” or nonmatching route where relevant; no invented account history is requested as proof.
| Question presented to a potential participant | Internal interpretation and exclusion rule |
|---|---|
| In the past six weeks, which parts of preparing a print job online have you personally done: upload an initial file, replace a file after seeing a proof, approve someone else's upload, something else, or none? | Continue when replacement was personally attempted. Approving only, interest without experience, or no qualifying activity does not match this round |
| Thinking about your most recent replacement attempt, approximately when was it and which service did you use? “Not sure” is acceptable | Check the six-week window and record current versus other-service context. Clarify uncertain timing; do not demand an order number or customer file |
| What happened on that attempt: completed without live help, stopped before completion, completed with live help, or not sure? | Assign one primary path using this attempt. Do not exclude stopped or assisted attempts merely to fill the calendar faster |
| In a sentence or two, what were you trying to change, and what did you do next? You may describe it generally | Check task fit without asking for proprietary details. An unclear answer needs clarification, not an automatic dishonesty label |
| Have you worked on ProofLane's design or this proposed research? | Exclude prior insiders for this round and record the reason. A job title elsewhere is not itself disqualifying |
| What participation format or arrangements would help you take part, and are any proposed times workable? You may discuss this separately | Handle logistics and requested support separately from experience eligibility. Missing arrangements mean unresolved readiness, not an irrelevant user |
A completed authored response shows how this works. P05 replaced a file in another proofing service two weeks earlier, stopped before the new file was accepted, and describes choosing a revised page layout before leaving the process. P05 requests keyboard-based participation. Mina classifies P05 as eligible in the stopped-attempt path, records the other-service context and the requested arrangement, and keeps diagnosis and customer documents out of the register. She does not classify P05 as confused about approval status; that has not been investigated.
Before using a real screener, check its wording and routing with appropriate reviewers and a small pilot. The example is a completed authored specification, not a screener that has been piloted with people. In particular, verify that a person can describe an unsuccessful attempt without being routed out as a non-user.

Compare the current list with the target
Mina's fictional October 27 register contains 12 response records. Seven meet the experience criteria, three do not, one needs clarification, and one is closed after the person declines further contact. Only four have confirmed logistical arrangements within the fictional case. That is not informed consent, attendance or completed research; all three remain separate.
The register below is the second completed asset. IDs stand in for entirely invented people; they are not anonymized real participants. “Confirmed” here means only that a proposed time and format were agreed within the story, with technical preparation still to be checked.
| ID | Most recent qualifying path or reason | Fictional disposition |
|---|---|---|
| P01 | Completed without live help; current service | Eligible, logistics confirmed |
| P02 | Completed without live help; other service | Eligible, logistics confirmed |
| P03 | Completed without live help; current service | Eligible, time unconfirmed |
| P04 | Stopped before completion; current service | Eligible, logistics confirmed |
| P05 | Stopped before completion; other service; keyboard arrangement requested | Eligible, logistics confirmed; technical setup still needs preparation |
| P06 | Completed with live help; current service | Eligible, support person's availability unresolved |
| P07 | Participated in ProofLane design | Excluded from this round for prior knowledge |
| P08 | Relevant attempt was outside the six-week window | Does not meet this round's recency criterion |
| P09 | Approved files but never attempted replacement | Does not match the task for this round |
| P10 | Timing and personal role unclear | Clarification needed; not counted eligible |
| P11 | Completed without live help; current service | Eligible reserve; no confirmed arrangement |
| P12 | Declines further contact | Closed; no further eligibility inference |
The resulting gap calculation uses unique people, not replies or invitations. Against the target of three independent completions, two stopped attempts and one assisted completion, the confirmed list contains two, two and zero respectively. The total is four of six, with one independent-completion place and one assisted-completion place still open. No research sessions have taken place in the fictional register; six completed sessions are still the plan.
P03 or reserve P11 might fill the first gap after arrangements are confirmed. Both match independent completion, so booking either would leave assisted completion uncovered. P06 remains eligible while the support arrangement is unresolved. Replacing P06 with an easier-to-book independent user would change the study's coverage, not complete the original target.
The two confirmed other-service participants also remain marked as such. Their other-service experience can inform the task discussion; it does not establish how current ProofLane users interpret its interface. Similarly, P05's requested keyboard arrangement does not mean all access needs are represented or that the study has evaluated accessibility.
Change the recruitment approach when a gap persists
Mina's completed recruitment review keeps the four logistical confirmations, retains P03 and P06 as pending, and leaves P11 as reserve. She records the two uncovered paths and asks the recruitment owner to examine approved channels that reach people who needed help. The proposed November9–13 window remains provisional. Mina records support availability and technical preparation as unresolved dependencies before confirming any final study schedule.
In a real project, existing opt-in research contacts may help reach current users, while an authorized professional community or recruiter may reach people using comparable services. Check permission and fit before contacting anyone. A customer-success list of satisfied customers is one channel, not a complete account of the intended population. If the criteria need to change, record the reason and how that change affects the research question before selecting more people.
Use the register to communicate the consequence of a remaining gap. If the team proceeds with only the four confirmed people, it can explore those participants' experiences but must state that assisted completion is absent. If that experience is central to the decision, defer or narrow the decision rather than imply that the absent group encountered no difficulty.
Collect only the information needed to manage recruitment and relevant experience. Keep contact details and operational arrangements in the organization's approved restricted process, separate from a broadly shared research summary. Ask what support would help participation, not for a medical explanation. Participant information and consent need their own process; an eligibility answer or calendar confirmation does not supply them.
Before the next study, compare your decision, screener and actual recruitment status side by side. Resolve one mismatch or write down the resulting limit before calling the sample ready. More workplace guides are available from LandOffer.ai; prepare a dated gap register showing eligibility, unconfirmed arrangements and the decision that remains limited by a missing perspective. Assign an owner to resolve or explicitly accept that limit.
Sources and Further Reading
- GOV.UK: Finding participants for user research, checked October 27, 2026; last updated September 22, 2026.
- GOV.UK: Write a recruitment brief, checked October 27, 2026; last updated March 31, 2016.
- Nielsen Norman Group: Screening Participants for User-Research Studies, checked October 27, 2026; published November 1, 2024.


