2learn

Recruiting participants

What it is

Recruiting is finding people who actually do the thing you are studying and getting them to give you an hour. It is the least glamorous and most common point of failure in research: a perfectly designed study that recruits colleagues, friends and the easiest possible users produces evidence about the wrong population.

Why it matters

The people you talk to decide what you can learn. Recruit only enthusiasts and every finding will be about power users; recruit only churned customers and you will design for people who already left. Recruiting well also sets the schedule: if you start recruiting the day the study is ready, you will have lost a week before the first session. And a bias that is invisible in your sample is invisible in your findings too — you cannot correct for it later.

Core ideas

Write the criteria before you look for people

Three to five screening conditions that describe the behaviour you care about, not the demographics. Then check the conditions with a screening question rather than an assumption.

Study: first-time buyer checkout
Screening:
1. Bought something online in the last 14 days
2. Used a payment method they had not saved before
3. Not employed in design, research or software
4. Not known to anyone on the team

Recruit through the behaviour, not the channel

Ask in the place the behaviour happens — a support queue, a follow-up email after a purchase, a community the users already belong to. A recruitment poster reaches people who like taking part in studies.

Screen out insiders and professionals

Colleagues, professional testers and people who know the product give you the answers they think you want. One screening question — “do you or does anyone in your household work in design, research or software?” — removes most of them.

Budget for no-shows and offer a fair incentive

Expect one in four to miss the slot. Offer something proportionate to the person’s time and constraints — cash, a voucher, or the organisation’s own product — and say in one line what the session involves.

Keep a small panel, not a big list

Five to eight people per distinct user group, reused across studies where sensible. A standing panel of people who match your criteria is worth more than a fresh anonymous list each time, provided consent allows it and they are not told what the study is looking for.

Practice

Write a screening script and a recruitment message

For your current question, write the criteria, one screening question per criterion, and a three-sentence recruitment message that states what the session is, how long it takes and what the incentive is.

Done when: a stranger reading only the message can tell in ten seconds whether they are eligible, without knowing what the study is testing.

Fill two sessions from a cold start

Run the recruitment you wrote with no internal mailing lists. Book two sessions from people who meet every criterion.

Done when: two people who are not known to the team confirm a slot, and you can say which screening question each of them passed.

Common mistakes

  • Recruiting people you know — they are helpful, informed and unrepresentative; fix: add “not known to the team” to the screening criteria.
  • Demographic screening only — age and location rarely predict whether someone does the behaviour; fix: screen on the behaviour and the constraints.
  • Recruiting at the last minute — the first session slips a week and the study loses momentum; fix: start recruitment the day the plan is agreed.
  • Over-representing enthusiasts — the sample is full of power users and the results read like a feature request; fix: include people who struggled, stopped or never started.
  • Explaining the study in the recruitment message — participants arrive primed and rehearse their answers; fix: describe the topic, never the hypothesis.

Go deeper

Worked examples

Analytics cookies are set only if you allow them. Nothing is loaded, and nothing is sent, until you choose.