UX and research · Beginner · 15 min
Recruit participants for user research
By Padding · Updated 18 Nov 2026
Any document editor, a spreadsheet and a calendar · November 2026
You will have a one-page recruitment plan, a private scheduling sheet and a first round of research sessions booked with the right people, including a spare for the person who does not turn up.
On this page
Most small teams do not fail at research because their questions are poor. They fail because they talk to whoever is easiest to reach: a colleague, a friend of the founder, the one loud customer who emails every week. Those people are real, but they are rarely the people the decision depends on, and what they say can send a design in the wrong direction with a lot of confidence behind it.
Recruiting is the dull part of research that decides whether the rest is worth doing. It is also the part with the least written about it for teams that have no research budget, no panel subscription and no operations person to chase people.
The running example here is a small team building class booking for a neighbourhood gym. They want to know why members who try booking online give up and phone the front desk instead. Swap in your own product and question as you go.
By the end you will have four things: a written description of who you need, a sourcing plan with more than one channel, a thank-you you can afford, and a scheduling sheet with five sessions booked plus one standby.
Step 1 · Describe who you need in behaviour, not demographics
Open a new document and call it "Recruitment plan". At the top, write the research question in one sentence. For the gym: "Why do members who start booking a class online stop and phone the desk instead?"
Under it, write who can answer that question. Describe what people have done, not who they are. "Women aged 25 to 40" tells you nothing about booking. "Has tried to book a class online in the last two months and phoned the desk at least once instead" picks out exactly the people with the experience you are asking about.
Write three short lists:
- Must have. Behaviours without which the session is wasted. Keep this to two or three items, recent enough to remember (the last one to three months).
- Mix of. Differences you want spread across the group so you do not hear one kind of story five times: new members and long-standing ones, people who book on a phone and on a laptop, people who visit at peak times and off-peak.
- Exclude. People whose answers will skew things: staff, their close family, anyone who works in product design or software at a competitor, and anyone who took part in your research in the last six months.
If you need to filter strangers against these lists, turn them into screener questions. That is its own skill, covered in the tutorial on writing a screener; here you only need the lists.
Check it worked
Your plan has a one-sentence research question and three labelled lists. Read each "Must have" item and ask: could I tell, from a yes or no answer about something the person did, whether they meet it? If an item names an age, a job title or a personality ("tech-savvy") rather than an action, rewrite it as an action.
Step 2 · Decide how many people and how much time
Numbers depend on the method, not on what feels respectable.
- Interviews about behaviour and needs. Plan five to eight per distinct group of users. Stop adding sessions when the last two told you nothing new.
- Usability tests of a prototype. Plan five per round, fix what you found, then test again with a fresh five. Two small rounds teach you more than one large one.
- Two or more very different groups (for example gym members and front-desk staff). Plan sessions for each group separately; their answers do not count towards each other.
Then add a margin for people who do not turn up. On a no-budget recruit, assume roughly one in five will cancel late or not show. For a plan of five, book six. For a plan of eight, book ten.
Write down the session length and format too. Shorter is easier to fill: 30 minutes on a video call is a small ask; 60 minutes in person at your office is a large one. Choose the shortest session that still covers your question, and say whether it is remote or in person.
Add a "Size and format" section to your plan with these numbers.
Check it worked
Your plan states a target number of completed sessions, a booking number that is higher than the target, a session length in minutes and remote or in person. If the booking number equals the target, you have no margin; add at least one.
Step 3 · List at least three places to find people
Relying on one source gives you one kind of person. Write a "Sources" section and list at least three channels, with a named person responsible for each and a target number from each. Pick from these, starting with the ones that cost nothing.
- Your own customer list. Existing users who match your "Must have" list. Ask whoever owns the customer data to pull a filtered list (for the gym: members with a failed or abandoned online booking in the last 60 days). Check what your privacy notice says about contacting customers for research before you send anything; if it does not cover this, ask your data protection lead first.
- In-product or in-person intercepts. Catch people at the moment they have the experience. A short banner or message after an abandoned booking, or a card at the gym's front desk that staff hand to people who phone to book. Intercepts find people who would never answer an email.
- Front-line staff. Receptionists, support agents and sales staff talk to your users every day. Ask them to note people who mention the problem and ask permission to pass on their details.
- Community groups. Local forums, neighbourhood chat groups, clubs, professional associations, course cohorts. Ask the group's organiser before posting; many groups ban unsolicited requests, and an organiser's introduction gets far more replies.
- Your own network, used carefully. Ask friends to forward your request to people who match, not to take part themselves. Second-degree contacts are fine; close friends will be too kind.
- Paid recruitment panels and agencies. Useful when you need a hard-to-find group quickly and you have money. Write this as a fallback, not a starting point, if you have no budget.
Next to each source, write the message you will send. Keep it under 100 words: who you are, what the session is about in plain terms (not your hypothesis), how long it takes, remote or in person, what they get as thanks, and one link or reply route to sign up. Do not mention the behaviour you are screening for in a way that tells people what to say to qualify; "We are talking to members about how they book classes" is better than "Have you ever given up booking online?"
Check it worked
Your "Sources" section lists three or more channels, each with an owner, a target number and a drafted invitation. Add the targets together: the sum should be at least three times your booking number, because most people you contact will not reply, and some who reply will not fit.
Step 4 · Choose a thank-you you can give
People are giving you their time, so thank them in a way that fits the ask. A participant who feels well treated is more likely to turn up, be candid and say yes next time.
Write a "Thank-you" section. If you have a budget, a voucher or a cash-equivalent gift sized to the session length is the norm. Agree the amount and how it will be paid with your finance contact before you send invitations, because promising something you then cannot pay damages trust fast. Some organisations have tax or anti-bribery rules for gifts, especially to public-sector staff or clinicians, so check first.
If you have no budget, there are still honest options:
- Something from the product. A free month, a class pass, priority access to a new feature. Check with whoever owns pricing that you can give it.
- A donation on their behalf to a charity they choose from a short list.
- Early influence. Some people take part to shape something they use every day. Promise a short summary of what you learned and what you changed, and then send it.
- Their convenience. Offer times that suit them, including evenings, and keep to the time you asked for.
Whatever you choose, state it in the invitation and give it within a few days of the session, not "when the project finishes". Never make the thank-you depend on saying positive things or on completing every task.
Check it worked
Your plan names one thank-you, who approved it, and the date by which each participant receives it after their session. If the answer to "who approved it" is blank, get that approval before the next step.
Step 5 · Set up a private scheduling sheet
Create a spreadsheet called "Recruitment tracker" and share it only with the people running the sessions. It holds personal data, so keep it out of shared team drives that everyone can open.
Add these columns:
- ID. A short code such as P01, P02. Use this in notes and recordings instead of names.
- Name and Contact. Email or phone, whichever they prefer.
- Source. Which channel they came from, so you can see whether one source is dominating.
- Fits. Which "Mix of" groups they cover (for the gym: new or long-standing, phone or laptop).
- Status. One of: invited, replied, qualified, booked, confirmed, done, no-show, cancelled, not a fit.
- Session. Date, start time and time zone.
- Consent sent and Thank-you sent. Dates.
- Delete by. The date you will remove their contact details, usually soon after the study ends. Agree it with your data protection lead.
Above the table, or on a second tab, write the session slots you can offer: specific dates and times, spread across mornings, lunchtimes and evenings, with at least 15 minutes between sessions for notes. Offer people two or three slots to choose from rather than asking "when are you free?", which turns into a long email thread.
Check it worked
Your tracker has every column above, sharing is limited to the named researchers, and there are at least as many open slots as your booking number. Fill in one dummy row (P00, status "not a fit") and check that each column has an obvious value to enter, then delete it.
Step 6 · Book, confirm and remind
Send your invitations. As people reply, mark them "replied", check them against your "Must have" and "Exclude" lists, and mark them "qualified" or "not a fit". Thank the ones who do not fit and tell them briefly that you have enough people with their profile this time.
For each qualified person:
- Offer two or three slots. When they pick one, send a calendar invitation straight away with the date, time and time zone, the length, the video link or the full address with directions, and a named contact with a phone number.
- Send your consent form or information sheet with the booking, so they can read it before the day. How to write one is covered in the tutorial on informed consent.
- Mark them "booked" and fill in the session column.
Watch the "Fits" column as you go. If your first five bookings are all long-standing members on laptops, pause that source and push the others until the mix looks right.
Reminders cut no-shows more than anything else. Send one message the day before asking people to reply to confirm, and a short one an hour or two before with the link or address. Mark them "confirmed" when they reply. Anyone who has not confirmed by the evening before gets a phone call or text, not another email.
Check it worked
Your tracker shows your booking number of people as "booked" or "confirmed", their "Fits" values cover every group in your "Mix of" list, and every booked person has a "Consent sent" date. Your calendar shows each session with a gap after it.
Step 7 · Plan for no-shows and late cancellations
Someone will not turn up. Decide now what happens so you are not improvising with an empty video call.
- Keep a standby. Your extra booking from the size step is your standby. If everyone arrives, run the extra session anyway or thank them and give the thank-you regardless; they kept their side.
- Wait ten minutes, then contact once. At ten minutes past, send one short message ("We are ready when you are, or we can find another time"). At fifteen minutes, mark them "no-show" and use the time to write up the previous session.
- Offer one reschedule. People miss sessions for ordinary reasons. Offer one new slot without pressure. If they miss that too, close their row politely.
- Track the pattern. After the round, look at the "Source" column for no-shows. If most came from one channel (often broad community posts), lower that channel's target next time and raise the more reliable ones.
At the end of the round, send each person who took part their thank-you and, if you promised it, the short summary of what you learned. Then delete contact details on each row's "Delete by" date, keeping only the IDs your notes refer to.
Check it worked
Your plan has a "No-shows" section with the wait time, the message you send and your reschedule rule. After the round, every row in the tracker ends in "done", "no-show", "cancelled" or "not a fit", every "done" row has a "Thank-you sent" date, and you have a calendar reminder on the earliest "Delete by" date.
Next steps
You now have a plan you can copy for every study. Keep the tracker template and reuse it, with a fresh copy each time so old contact details do not linger.
- Turn your "Must have" and "Exclude" lists into short questions with Write a screener survey to find the right participants.
- Prepare the information sheet and consent form you send with each booking in Get informed consent for user research.
- Use your booked sessions in Run your first user interview or Plan a usability test.
Part of these paths
Related tutorials
- Write a screener survey to find the right participantsYou will have a screener of eight questions that finds people who booked a clinic visit in the last six months without hinting at the answer you want, plus written rules for who qualifies, who is turned away and who fills a quota.
- Get informed consent and store research data safelyYou will have a plain-language consent form for one study and a data-handling checklist that says what you record, where it lives, who can see it and when it is deleted.
Need this done for you?
Padding does this work for growing companies, from the first diagnosis to a system that keeps working.
Get new tutorials