How to recruit user research participants: your users or a panel?
Recruitment determines whose experiences your study can describe. A well-written question cannot recover context from someone who has never done the relevant work. Before choosing a participant source or a target response count, describe the experience people need in order to help you answer the research question.
For product teams, a common choice is between existing users and an external participant panel. Both can be useful. The right choice depends on whether you are investigating your product’s history, a broader workflow, or the first encounter with something new. Here is a practical way to make that decision.
Define participants through relevant experience
Replace a broad audience such as business users with a description tied to the question. For a hypothetical report-sharing study, that might be people who prepared a recurring report and shared it with another team within the past month. If the study concerns your current permissions model, specify that participants have managed access in your product.
Separate requirements from useful variation. Recent report sharing might be essential; company size and job title might be dimensions to compare rather than reasons to exclude someone. An unnecessarily narrow screener can remove informative participants and make recruitment harder without improving the evidence.
Also identify people whose experiences are easy to miss: occasional users, people who needed help, and people who stopped before completing the workflow. If your recruitment channel cannot reach them, record that gap in the study plan. Their absence is part of how later findings should be interpreted.
When to recruit from your own users
Recruit current users when the question depends on familiarity with your specific product, account setup, or recent experience. They can describe how the product fits into their work and what happened before or after a particular action. An external participant can try an interface, but cannot supply a history of using your product that they do not have.
Existing users also bring learned behavior. A long-time administrator may know that an ambiguously named setting controls access because they encountered it months ago. If you want to understand first-use comprehension, include newer users or people who have not performed that task before. Distinguish product expertise from expertise in the underlying job.
In-app recruitment reaches people who are present in the product. That is useful for contextual questions, but it creates a boundary: people who no longer visit, cannot sign in, or abandoned an earlier step may never see the invitation. Do not treat that audience as a complete picture of everyone affected.
When an external participant panel fits
A panel can help when you need people beyond your current user base, such as prospective users with relevant workflow experience. It can also be useful for evaluating how an unfamiliar person approaches a task. Start with the job and context the person needs to understand, then check whether the available criteria and screening questions can identify that experience.
External does not mean representative. A panel is a recruitment source, and its fit still depends on eligibility, coverage, willingness to participate, and the study itself. Avoid presenting panel responses as a market estimate unless the sampling and analysis support that claim. A qualitative round can reveal useful explanations without measuring how common they are.
Palette’s user studies and usability tests support recruitment from your own users or its external participant panel. Microstudies collect answers from your own users within the product.
Write a screener that checks experience without coaching
A screener should establish eligibility. It should not tell people which answer is most likely to qualify them. “Are you an expert administrator who struggles with permissions?” both advertises the desired answer and combines expertise with the problem you hope to investigate. Ask about responsibility and behavior separately.
| Need to establish | Example screener question | What to watch |
|---|---|---|
| Recent activity | When did you last share a recurring report with another team? | Use a time window relevant to the study |
| Responsibility | Which parts of preparing and sharing reports do you personally handle? | Distinguish doing the work from observing it |
| Product familiarity | Which of these tools have you used for that task? | Do not reveal a preferred answer in the invitation |
Review the full recruitment journey, including the invitation, screener, study instructions, and any reward terms. Someone who qualifies on paper may still misunderstand the task or be unable to complete the format. A pilot helps you catch that mismatch before you interpret short or irrelevant responses as lack of interest.
Decide response targets around the study purpose
There is no participant count that makes every research question adequately answered. Think about variation in the audience, the depth of the questions, the method, and whether you are exploring explanations or estimating a quantity. Do not use a small exploratory sample to announce a precise percentage for your entire customer base.
Before launch, decide what would make you run another round: missing roles, conflicting explanations, unusable responses, or a newly discovered audience. That rule is more useful than treating the initial target as a guarantee. If you compare groups, keep track of how much evidence you have from each one rather than relying on the combined total.
After collection, include recruitment source, eligibility criteria, relevant gaps, and study dates in your research note. The continuous research guide shows how to vary recruitment across rounds. The study question guide helps you connect the audience you recruit to questions they can actually answer.