Imagine spending three weeks recruiting participants for a usability study of your mobile banking app. On test day, you discover that half of them have never used mobile banking. Two only bank in person. One signed up because the incentive looked good.
The sessions still happen. The data still comes in. But none of it can be trusted, because the people behind it were never your users in the first place.
This is the most expensive mistake in UX research, and it happens before a single research question is asked. The fix is a well-built screener. This guide covers what a UX research screener is, which questions belong in one, how to avoid the traps that let the wrong people through, and how to build one quickly with a tool like Formplus.
A UX research screener is a short questionnaire that filters potential participants before a study begins. It usually runs 5 to 10 questions and takes under three minutes to complete. Its only job is to answer one question: does this person match the profile we need for this research?
A screener is not the study itself. It sits in front of the study, like a bouncer at a door. People who match your criteria get invited in. People who don’t are politely turned away before they cost you time, budget, or bad data.
Screeners are used before almost every type of UX research, including:
It is tempting to treat screening as admin work. It isn’t. The quality of your participants sets a ceiling on the quality of your findings, and no amount of clever analysis can raise that ceiling later.
Here is what a good screener actually buys you.
Valid data. Feedback from real users of your product, or people who genuinely face the problem you are studying, reflects reality. Feedback from anyone else reflects guesswork.
Lower cost per insight. Most teams pay incentives per session. If two out of eight participants are unqualified, a quarter of your incentive budget and your researcher’s time produced nothing. Screening moves that spend to people who can actually inform decisions.
Faster recruitment. This sounds backwards, but it’s true. A clear screener with automatic disqualification logic filters hundreds of applicants without anyone reviewing them by hand. Manual vetting is what slows recruitment down.
Better product decisions. Research exists to reduce risk in decisions. Decisions built on input from the wrong people don’t reduce risk. They hide it.
Every weak screener has the same root cause: the team never agreed on who they were looking for. Write the participant profile first, then write questions that test for it.
A useful profile answers four questions:
Notice that this profile is built on behavior, not identity. “Has transferred money on mobile in the last 30 days” tells you far more than “aged 25 to 40.” We will come back to this.
One more tip: write the profile down and get sign-off from your stakeholders before recruiting. It takes ten minutes and prevents the awkward mid-study conversation where a stakeholder asks why “none of these people are our customers.”
Most screeners draw from six question types. You won’t need all six every time, but you should be able to justify every question you include.
1. Role or occupation. Confirms professional context when it matters, and catches industry insiders when it doesn’t. Example: “Which of the following best describes your current occupation?” Include “UX design or research” and “Market research” as hidden disqualifiers.
2. Product or category experience. Confirms they use the product, or a competitor, or the category you are studying. Example: “Which of these apps have you used in the past 3 months? Select all that apply.” List your app among competitors and one fake app. Anyone who selects the fake app is disqualified for careless or dishonest answers.
3. Frequency of behavior. Separates active users from people who tried something once. Example: “How often do you shop for groceries online?” with options from “Weekly or more” to “Never.”
4. Recency of behavior. Memory fades fast. If your interview relies on someone recalling an experience, that experience should be recent. Example: “When did you last book a flight online?” A participant who booked last week gives you detail. A participant who booked two years ago gives you reconstruction.
5. Technology and setup. Confirms they can actually take part, especially for remote tests. Example: “Which of these do you have access to for a 45-minute video call? A laptop or desktop with a working camera and microphone.”
6. Availability and consent. Confirms scheduling fit and willingness to be recorded, so no session collapses at the last minute.
A screener built from these six types typically lands at 6 to 9 questions. If yours is much longer, something in it isn’t earning its place.
Behavior, almost every time.
Demographics describe who a person is. Behavior describes what a person does. UX research is the study of what people do, so behavioral criteria predict participant quality far better than age, gender, income, or education.
Consider a study of a budgeting app. A 22-year-old student who tracks every expense in a spreadsheet is a far better participant than a 45-year-old accountant who never budgets personally, even though the accountant “looks” more qualified on paper.
That said, demographics are not useless. Use them in two situations:
The rule of thumb: screen in with behavior, balance with demographics. Never use a demographic filter you cannot connect to your research goals, because every unnecessary filter shrinks your pool and slows recruitment.
Screeners fail in a specific way: they telegraph the “right” answer, and applicants who want the incentive simply give it. Your job is to make the qualifying answer invisible.
Never reveal what you are looking for. Compare these two questions:
Avoid leading language. “How much do you enjoy shopping online?” assumes enjoyment. “How would you describe your experience with online shopping?” doesn’t.
Ask one thing per question. “Do you use fitness and nutrition apps?” is two questions wearing one question’s clothes. Someone who uses only fitness apps cannot answer it honestly.
Prefer specifics over self-assessment. People overrate themselves. Instead of “How tech-savvy are you?”, ask “Which of the following have you done in the past month?” with concrete tasks like paying a bill online or installing an app.
Keep answer options mutually exclusive and complete. If your frequency options are “1 to 3 times” and “3 to 5 times,” a person who did something three times has two homes. Always include an “Other” or “None of these” escape hatch.
Even experienced teams fall into these traps.
Making the incentive answer obvious. If your screener reads like a checklist for a reward, professional survey-takers will pass it. Randomize answer order, hide your criteria among decoys, and use a fake-brand trap question.
Screening only on claims, not evidence. “Do you have experience with project management tools?” invites a yes. “Which project management tool did you use most recently, and what did you use it for last week?” invites proof. Open-text verification questions are the single best filter for dishonest applicants.
Forgetting to screen out insiders. Designers, researchers, marketers, and employees of competitors behave nothing like normal users in a session. Always include an occupation question with these as silent disqualifiers.
Over-filtering. Every extra criterion multiplies recruiting difficulty. If you need people who use your app daily, on Android, in a specific city, aged 30 to 35, who switched from a specific competitor, you have designed a study for about nine people on Earth. Rank your criteria as must-have or nice-to-have, and screen only on the must-haves.
Writing a screener that is really a survey. If your screener is collecting opinions, attitudes, and feature wishlists, it has stopped screening and started researching, and it is doing both badly. Keep it lean. Save the real questions for the study.
A screener only works if people finish it. Drop-off in the screener silently shrinks and skews your pool, because the people most likely to abandon a long form are often the busy, real-world users you most want.
Keep it under three minutes. Six to nine questions is the sweet spot. Every question past ten costs you completions.
Lead with easy questions. Start with simple selections. Put open-text and consent questions near the end, once people are invested.
Use disqualification logic early. If someone fails a knockout criterion at question two, end their screener politely right there. It respects their time and keeps your data clean. This is where skip logic in your form tool earns its keep.
Be upfront about the process. State the study length, the incentive, and the timeline on the first screen. Vague screeners feel like scams and get abandoned.
Make it mobile-friendly. A large share of applicants will open your screener on a phone. If it breaks there, they are gone.
Close the loop. Tell disqualified applicants clearly and kindly that they didn’t match this study’s profile, and invite them to a panel for future research. Today’s screen-out is next quarter’s perfect participant.
You pilot your studies. Pilot your screener too. Send it to 5 to 10 colleagues or a small slice of your audience before full launch, and check four things:
Then keep monitoring after launch. If 200 people have applied and four have qualified, don’t wait for the miracle. Loosen a criterion and relaunch.
Everything in this guide comes down to three practical needs: hiding your criteria, disqualifying automatically, and keeping completion friction low. This is exactly what a capable form builder handles for you.
With Formplus, you can:
You can start from a blank form or adapt one of the existing Formplus survey templates, and have a working screener live in under an hour.
A UX research screener is a small artifact with outsized consequences. It is the difference between a study built on your real users and a study built on strangers who wanted a gift card.
The playbook is short. Define your ideal participant in terms of behavior. Ask only questions that test for that profile. Hide your criteria so they can’t be gamed. Verify claims with specifics. Keep it under three minutes, pilot it before launch, and adjust when the numbers tell you to.
Do that, and every hour of research that follows is built on solid ground. Ready to put it into practice? Build your first screener free with Formplus and start recruiting participants you can actually trust.
You may also like:
Getting free access to academic articles is one of the best things that can happen to you in your research journey. However, over 10,000...
In this article, we are going to discuss multistage sampling, its uses, the advantages, and the disadvantages.
One way to visualize the similarity between variables in statistical data analysis is by stacking data points (or dots) in a column,...
In this post, we will focus on what a focus group definition is and provide tips for conducting a successful focus group.