USER RESEARCH
12 simple techniques for user interview questions
Formulating questions that foster more reliable answers
Aseasoned interviewer takes care of many things: builds hypotheses, chooses interviewees, schedules appointments, sets the stage, writes the script, and so on. Any of these preparations can go wrong, but a bad script undermines all efforts.
Originally published in Smashing Magazine.
![](https://miro.medium.com/v2/resize:fit:700/1*ZHJOXla4N1u9NMsPE9AU7Q.png)
If you are a designer who hasnât interviewed users before and there is no dedicated UX researcher nearby, aim at crafting robust questions. Then, there is a chance theyâll smooth out possible shortcomings. Right questions donât simply roll off the tongue, but itâs a handy skill one can train. Today weâll focus on problem interviews that help capture a real-life problem your digital product is going to solve. You might already have a solution or its prototype, or you may want to build one.
Pitfall 1. Hypothetical questions
âI donât care if people will use our new features,â said no budget owner ever. Investing in design and development, people want to make sure money will return. And direct asking, unfortunately, is not an effective way to check it, although it may intuitively seem a great idea.
![](https://miro.medium.com/v2/resize:fit:700/1*xIIG-whCEaIH9XhVlEkGkg.png)
In my practice, there were a lot of cases when people said they liked something but were reluctant to pay for it or utilize it later on. So, is there any method to make sure that a product or separate feature will be of high demand when implemented?
![](https://miro.medium.com/v2/resize:fit:700/1*_IPrxlG4J1ip7utHeaY9LQ.png)
I cannot recall anything more relevant than referring to peopleâs experiences and examples of similar behavior from the past. For example, if your users donât have a habit of saving articles for later in all their news apps, what is the chance theyâll start doing it in your app? As Jakob Nielsen said, âUsers spend most of their time on other sites.â
Letâs recap:
- Donât overuse the questions like âwould youâŠâ âhow probable is thatâŠâ and âwhat if there wereâŠâ
- Try to find out if a person has relevant experience.
Pitfall 2. Closed questions
Closed questions appear from a natural human wish to be approved and gain support. However, in the interviews, they arenât that useful. A yes-or-no question doesnât provoke reserved people to talk and doesnât help much to reveal their motives and way of thinking.
![](https://miro.medium.com/v2/resize:fit:700/1*jxXKVLm7ej9okDXTfhq0pg.png)
To be fair, closed questions are not evil. For example, they can serve a handy facilitation technique to make a talkative interviewee stop and turn back to the point. Also, they can help to double-check the information previously received through open questions. But if your goal is to gather as much data as possible, open questions will work better.
![](https://miro.medium.com/v2/resize:fit:700/1*iNdPR1b0FNaZHXlNZVsOBQ.png)
Letâs recap:
- Aim at making your interviewee tell stories, not approve/reject variants.
- Hot-fix closed questions with âhow.â Instead of âDo you do X often?â say, âHow often do you do X?â
Pitfall 3. Leading questions
The things considered polite in everyday conversations may be harmful to the efficiency of a user interview. Trying to help an interviewee with the options can guide them in saying what they donât really think.
![](https://miro.medium.com/v2/resize:fit:700/1*SJWoficHfhyew0UvfbnZNA.png)
A user interview is not the most comfortable situation for the majority of people, and they try to pass it as quickly as possible and at minimum effort. As a result, people tend to agree with anything more or less close to the truth or with a socially-expected choice instead of composing their answer from scratch. Thatâs why itâs better to move one step at a time and build the next question upon the answer to the previous one.
![](https://miro.medium.com/v2/resize:fit:700/1*0wmX_tfRiu1v2ffjFPgbkg.png)
Letâs recap:
- Donât try to help your interviewee â itâll discard answer reliability.
- Better rephrase a question than suggest answer options.
Pitfall 4. Selfish questions
Idea authors sometimes act like proud parents â they want everyone to admire their child. The downside of such an attitude in user interviews is the unconscious usage of the pronouns âweâ or âour.â As a result, users feel as if they are taking an exam and should either adore what they see or maintain neutrality, disguising real complaints.
![](https://miro.medium.com/v2/resize:fit:700/1*APnRcqvnJejxSSvhGetKTQ.png)
In your interview script, replace possessive pronouns with neutral words like âthis siteâ and âthat applicationâ or just call a subject of conversation by the name. Pro tip: as an interviewer, you can try hiding or understating your title and relation to the topic. For instance, before discussing digital products, donât emphasize you are a designer.
![](https://miro.medium.com/v2/resize:fit:700/1*GcUTygGCeIvsfLi1f2ir2A.png)
Letâs recap:
- An interview is not a demo. Donât show your relation to the topic.
- Edit âI,â âwe,â and âourâ out of the question list and as much as possible from other materials shared with interviewees.
Pitfall 5. Stacked questions
There are many reasons why we ask stacked questions. It can be a human desire to be heard, the fear of being interrupted, or worrying that you might forget the next question while listening to the current answer. However, for interview efficiency, stacked questions are not an option. Interviewees often select the one they are more comfortable to answer to or the one they managed to memorize from the stack.
![](https://miro.medium.com/v2/resize:fit:700/1*l-IPgwPOf_90GEXiPtEp0w.png)
Remembering questions shouldnât become the intervieweeâs burden, so itâs better to ask them one by one. And maybe the answers are so comprehensive that you wonât need some of the planned questions anymore.
![](https://miro.medium.com/v2/resize:fit:700/1*7L7OZcntP4dnKXK30gG7Fw.png)
Letâs recap:
- Ask one question (sentence with a question mark) at a time.
- If you feel you need to ask about several things at once, better compose and ask one broader question.
Pitfall 6. Explanation instead of a question
Teams that work together for some time often establish own language and tend to bring it into the product they are building. But will users understand such words as âdashboard,â âmodule,â âinclusion,â or âtriggerâ? Explanatory questions put an interviewee into the position of a lexicographer and check what sense (if any) they put into brand concepts and expert terminology. For a designer, it gives insights into how the future product â a website, app, or self-service terminal â should speak to people.
![](https://miro.medium.com/v2/resize:fit:700/1*23MFJUuYJIC2cpFK5fw7Og.png)
The opposite side of this approach is explaining it yourself and leading people before they have a chance to share their opinions. Think about this: in the interview, you are superior and can put pressure on users making them get your point. But will you always be there for thousands of users to explain how the product works? Probably no. So, itâs more efficient to discover peopleâs thinking styles and then create self-explanatory solutions rather than create something and push it in interviews.
![](https://miro.medium.com/v2/resize:fit:700/1*u6Z82HIm_0Yhy4QgSGaVGw.png)
Letâs recap:
- An interview is not a pitch: interview âsalesâ are fake sales.
- If you feel an interviewee might be out of context, you either invited the wrong people or want to defend an unclear solution.
Pitfall 7. Question clutter
Open questions are great until you realize there are too many details to figure out. The best method in such a situation is storytelling â describing a recent or the most prominent experience. As a result, an interviewee talks about a real situation and is less inclined to compose a socially desired answer or summarize various cases.
![](https://miro.medium.com/v2/resize:fit:700/1*CYOlOYLagI77EK7gwsIsxg.png)
Besides, storytelling gives the freedom to speak about aspects a person considers necessary. Usually, people start with or talk longer about the most crucial experiences.
![](https://miro.medium.com/v2/resize:fit:700/1*aqE_wMZu8yzaDZUNMLNeag.png)
Letâs recap:
- When in doubt which aspect to start with, ask to tell the whole story.
- Sometimes an optimal interview question is not a question.
Pitfall 8. Too theoretical questions
When youâve figured out regularity or general attitude, itâs the right moment to ask the interviewee about an example. Recent-experience questions can fill in the gaps, which might have appeared while answering general questions.
![](https://miro.medium.com/v2/resize:fit:700/1*tV-LBoqz1s4OwSFzxno5EQ.png)
For an interviewer, itâs another powerful method to check if users arenât accidentally exaggerating or dropping significant details.
![](https://miro.medium.com/v2/resize:fit:700/1*lL-TCRJj6w_Muy8PdPgPNA.png)
Letâs recap:
- Encourage your interviewee to illustrate all significant thoughts with recent or most remarkable examples.
- Try to find the real-life equivalents of the metrics you want to ask about (for example, interest â reading time).
Pitfall 9. Talking about what you can observe
If you are lucky enough and interview people in their ânatural habitat,â itâs a perfect chance to see their work process with your own eyes. So, if there is an opportunity to ask a user to demonstrate typical actions â offline or online â youâll gather tons of insights.
![](https://miro.medium.com/v2/resize:fit:700/1*WsWeSlZ0ViE-x7f3YsPkwA.png)
Itâs a chance to learn about usersâ habits (including shortcuts and favorite programs), level of computer skills, software environment, and the way of thinking (mental model).
![](https://miro.medium.com/v2/resize:fit:700/1*FTsKvepuT60vTo_RSjoKSA.png)
Letâs recap:
- Whenever possible, ask interviewees to show it rather than describe it.
- Prepare a compelling explanation of why you need to see something with your own eyes; otherwise, you may get rejected.
Pitfall 10. Tolerating vagueness
Abstract nouns and adjectives, for example, âcomfort,â âaccessibility,â âsupport,â âsmart,â or âuser-friendly,â are probably the trickiest words in the language because everyone interprets them differently. When you hear abstract names, thatâs not enough to document them as they are. These words require âunboxingâ and only then can support design decision-making.
![](https://miro.medium.com/v2/resize:fit:700/1*OYp630Z363_huK_IMNRc8w.png)
Want to hear my interview mantra? âNothing is clear enough.â Itâs my second favorite slogan after the classical UX phrase âIt depends.â It means that you cannot be certain about the meaning of intervieweeâs words if you hardly visualize a scenario behind them. The best way to unbox abstract concepts is by turning them into verbs.
![](https://miro.medium.com/v2/resize:fit:700/1*MOw-8C0S0k2konT0yluZHA.png)
Letâs recap:
- Abstract nouns and adjectives are not helpful and actionable insights.
- Donât hesitate to unbox abstract words until you are able to imagine the scenario of what was happening.
Pitfall 11. Overlooking exaggerations
Generalizations like âall,â ânever,â âalways,â ânobody,â âoften,â or âfrequentlyâ are as unclear as abstract nouns and adjectives. But the way to âunboxâ generalizations is different â through quantifying. Basically, you ask questions about approximate numbers or proportions.
![](https://miro.medium.com/v2/resize:fit:700/1*6UZxkuTkJjTjqCMCGz1sDw.png)
An interviewee, of course, might not provide you statistics, but at least youâll understand whether the userâs âvery frequentâ is about âmore than a halfâ or ânearly 20%.â Another example: the same phrase âa lotâ can mean â50 per dayâ for emails but only â5 per yearâ for critical cybersecurity alerts.
![](https://miro.medium.com/v2/resize:fit:700/1*jf1w2EE4r3YkElReow3WaQ.png)
Letâs recap:
- When you hear vague characteristics of frequency, probability, or quantity, donât forget to figure out what they mean to a person.
- To unbox a vague adverb, ask about an approximate number behind it.
Pitfall 12. Undervalued WH-questions
As a non-native speaker, I remember these questions from the English classes at school. The teacher often asked us to make WH-questions (What? Where? When? Who? How?) so that we could start a conversation and break the awkward silence. Nothing had changed from school times. Now, as a designer, I often use WH-questions as the main interviewing instrument.
![](https://miro.medium.com/v2/resize:fit:700/1*hRrN6MmLG5YW13N2wzIg4Q.png)
My favorite question is âwhy.â For the sake of politeness and a more friendly atmosphere, I conceal it behind the following phrases, âWhat are you trying to achieve when youâŠ?â or âCan you please explain the reason/value ofâŠ?â This is how in pursuit of a root cause you can ask several âwhysâ in a row without annoying your interviewee.
![](https://miro.medium.com/v2/resize:fit:700/1*9qUQplnqKcyd1ZPgAnkWCg.png)
Letâs recap:
- WH-questions help to understand all circumstances: who, what, where, when, why, and how.
- A good interview script usually includes a lot of WH-questions.
Summary
The question techniques above are pretty straightforward and might not take into account the nuances of a particular conversation or interviewee. Of course, even the best questions wonât make all the answers automatically objective, but they can make information more reliable and actionable. All in all, itâs always on an interviewer to adjust according to the situation. Here are the three core principles if you are in doubt about particular questions.
Experience holds more truth than a hypothesis
Thatâs why itâs recommended to ask about cases from the past and similar examples from other areas of a userâs life.
Let them tell their story; your ideas can wait
The goal of an interview is to explore the truth. If you force one interviewee to support you, it doesnât mean youâll cope with persuading a thousand. Also, give preference to clarifying the unknown versus checking hypotheses â for hypotheses, a better method is prototyping and testing.
If you cannot imagine it, you donât get it
In a series of 1â2-hour user interviews, itâs so easy to get lazy and pretend you understand what you hear. Try challenging the intervieweeâs statements in your mind, âDid he say the truth? Do I know why she says that? What exactly do they mean telling me about it?â
![](https://miro.medium.com/v2/resize:fit:700/1*5ZLPS0bNAkhK1sDhIgbNsg.png)
Bonus: self-check exercise
Take a sheet of paper and write numbers from 1 to 15 in a column. Read the following question examples. If you consider a question formulated properly for an interview, put a âplusâ (+) in front of the number. If you think it wonât work, put a âminusâ (â). Correct answers are after âRecommended reading.â
- What are your work responsibilities?
- What are your main challenges with our web application?
- How do you start your working day?
- Do you usually publish just text or text with photos?
- What do you feel when the button is not working?
- Please tell me about your company: its business model, primary markets, KPIs, and organizational structure.
- Do you like the autosave feature?
- How do you use the âSmart Publishingâ module?
- What are you trying to achieve by archiving drafts?
- Would you click on the âGet startedâ button if we add the note that the trial doesnât require a credit card number?
- What do you think this button does?
- Youâve just said âuser-friendly.â Please tell me more about the apps you use and consider to be user-friendly?
- How much will you pay for the ability to watch and restore previous versions of a document?
- What do you take into account to mark a project âApprovedâ?
- On which of the stages that youâve just mentioned do you have the most critical issues?
Recommended reading
- âThe Mom Test: How to Talk to Customers and Learn If Your Business is a Good Idea when Everyone is Lying to You,â a book by Rob Fitzpatrick.
- âUser Interviews: How, When, and Why to Conduct Them,â an article by Kara Pernice for Nielsen Norman Group.
- âThe 3 Types of User Interviews: Structured, Semi-Structured, and Unstructured,â a video by Maria Rosala for Nielsen Norman Group.
- âThe Art of the User Interview,â an article by Nick Babich for Springboard.
- âHow to Conduct User Interviews,â an article by Interaction Design Foundation.
- âFirst Rule of Usability? Donât Listen to Users,â an article by Jakob Nielsen for Nielsen Norman Group.
- âInterviewing Users,â an article by Jakob Nielsen for Nielsen Norman Group.
- âThe Art of Asking Good Questions with The Language Compass,â an article by Stefano Mastrogiacomo.
Self-check: correct answers
Interviewing is not a mathematically precise topic, so please donât take the following answers as dogma. However, they can serve as a benchmark and help you to enhance your next interview script.
- + (Open question.)
- â (Has the word âour,â which is selfish.)
- + (Open question, a good conversation starter.)
- â (Provides options, a guiding question.)
- +/â (Open question; however, feelings are subjective and might not help to discover what a person tries to accomplish.)
- â (Stacked questions.)
- â (Closed question; has the word âfeature,â which belongs to IT terminology.)
- â (Includes the branded name of a feature, a selfish question; but if users are familiar with this feature, it might not be a problem.)
- + (Open question about the way of thinking.)
- â (Hypothetic question.)
- + (Open question about the way of thinking.)
- + (Past-experience and unpacking question aimed at understanding what an abstract word means to an interviewee.)
- â (Hypothetic question.)
- + (Open question about the decision-making style.)
- + (On which stage = when; WH-question, although doesnât sound like one.)
- If you found this article useful, feel free to buy me a coffee â
- Here is a reading list with all my design articles on Medium.
- Letâs connect on Linkedin and Twitter.
- Drop me a note, if you want me to present this or another design-related topic at a conference or meetup.
- Bored by âhow-toâ design articles? Check my other blog where I talk about overlooked architecture.