The Complete Guide to User Interviews
“Life’s too short to build something nobody wants.”
― Ash Maurya, Running Lean
I am writing this guide for those who want to conduct user interviews and need help figuring out where to start and how to get maximum results.
In a nutshell, the user interview process looks like this:
1. Define what we want to learn from our users (hypotheses)
2. Choose the type of user interviews
3. Write a Research Plan document and align it with stakeholders
4. Recruit participants
5. Conduct user interviews
6. Analyse user interviews
7. Share research insights with stakeholders
For many companies, the first step in planning user interviews will be not writing hypotheses but convincing stakeholders that we need user interviews.
So let's start from the very beginning…
Why do we need user interviews?
One of the technics conducting user interviews is to ask endless "Why" questions. It helps to dig deeper into real user problems and ignore generic answers telling us nothing.
Let's imagine that we are working on a new product and use this "endless why" method to answer our question:
- Why do we need user interviews?
Because we want to test a new idea before creating MVP. - Why do we want to test a new idea?
Because we are curious if our users need it. - Why do we need to be sure that our users need it?
Because the implementation of this idea takes three months. - Why is that important for us?
Because we don't want to lose time and money. - Why is that important to us?
Because we want our business to be profitable, and it is only possible if we solve our customers' problems and they are willing to pay for it. If nobody needs our product, we run out of budget and fail as a business.
As you can see, the purpose of user interviews is to save time and money for business and provide opportunities for growth by defining user problems that matter.
What do we want to learn?
To get the most results from user interviews, we need to define What we want to learn from our users. The best way to do that is to write hypotheses.
Here are some tips:
- Keep your hypotheses short and precise.
- Invite stakeholders to create hypotheses together. It will help to be aligned and develop trust when you share research insights with them.
Choosing the type of user interviews
We can choose the right type of user interview based on what we want to learn from our users.
Generative interviews
This type of interview helps us to answer the question: "What I don't know?."
You don't need a strict script for generative interviews but only a few open-ended questions to start the conversation. So the only thing you need to do here is to listen carefully and ask follow-up questions.
Continuous interviews
Continuous interviews are interviews you do regularly to validate your assumptions. For these interviews, you need to have a list of questions based on created assumptions.
User interviews vs. Usability tests
It's easy to confuse user interviews with usability tests. The main difference is that the usability test focuses on design (early sketch, prototype, or working software) and observation. During the usability test, you will learn (and not learn during the user interview) the following:
- Whether a design is easy to use
- What makes a design easy or difficult
Tip: If the user starts to share insightful thoughts during the usability test that might improve your product — don't hold yourself and ask him to tell you more about it.
In my practice, it happened when the usability tests became generative interviews. The goal is to find users’ problems, needs and values. If you feel like you are on the right track to finding them — don’t stick to the rules and mix interview types. Improvise!
Write Research Plan Document
The next step in planning user interviews — write a Research Plan document. This document includes the following:
- Background: Why is this research initiative important from the users' perspective? Will it help us uncover new mental models or journeys? Or pain points that lead to a more satisfactory user experience? Will it teach us how to iterate on current designs?
- Internal stakeholders: List stakeholders involved in this project and their associated roles.
- Research objectives: You should have 3–5 objectives for your product, and they should be focused on the following:
- understanding users' thoughts, needs, and perceptions
- exploring new ideas/concepts
- improving the current experience
- evaluating the performance of a product - Business objectives: Why is the business interested (or why should they be interested) in running this research project? What is the problem you are trying to understand? What financial KPIs could this impact (think acquisition, revenue, retention)?
- Methodology: Based on the goals/objectives, what methods will we use? Is this project more generative, calling for 1x1 research interviews? Or more evaluative-focused, which would move towards usability testing? How long will the sessions be? How many will there be? Will they be remote or in-person? What tools will we use? What devices?
- Metrics: Based on the methodology, decide on different metrics for sessions like usability testing, including time on task, task success, ease of use, etc.
- Participants: Are there any participant qualifications that need to be met? For example, do they need to be a specific type of user? Is it necessary that they have browsed your website or have yet to use your app? Is there a demographic mix (gender, location, income, age)?
- Anticipated timeline: Give an anticipated timeline of how long the research will take, from recruiting through analysis and sharing results.
- The interview guide: Introduction, script, wrap-up.
- Resources: Additional links relevant to the project, such as Jira or previous studies. Once complete, links to the videos and research summaries.
Recruit participants
After you understand what type of users (persona) you are looking for, you can begin to recruit them.
If you still need to get a user database, it's time to create it. A user database will help you save time searching for the right people. All you need to do is to contact them and invite them for interviews.
Surveys work well for this purpose. Based on the personas or types of users you are looking for, create questions for the survey and post this survey on social media or send it by email to your users.
Check out my article about user personas:
Conduct user interviews
Here are practical tips that helped me to conduct user interviews:
- Spend the first 5 min creating a friendly atmosphere. Introduce yourself and share the purpose of this interview. Ask the participant to tell about himself. Smile :)
- Tell the participant that there are no right or wrong answers. It is essential that the participant feels comfortable sharing any thoughts without thinking it is wrong or right.
- Ask if it is ok to record the interview. In most cases, people agree to record interviews. You can share videos from interviews with stakeholders. It is also helpful for notetaking purposes — if you are conducting an interview alone, you don't have much time to take notes. That is why it is better to conduct interviews with a partner. One person is talking, and another is taking notes.
- Ask open-ended questions. Don't ask questions that can be answered by "yes" or "no." Instead, the question should motivate users to talk.
- Ask follow-up questions. The first answer rarely gives you an understanding of the problem. So dig deeper and ask the participant to clarify it for you.
- Use questions "Why?" and "Why is that important to you?". I've got lots of insights just by endlessly asking those questions :)
- Repeat the last exact words that the participant said and pause… This works like magic. People don't like silence…they want to fill it. So they always start talking. You will feel uncomfortable when you start doing that but hold on until the participants start talking. They will try to explain what they just said, and you will be surprised how it is different from the first answer they gave you. I learned it from the book Never Split The Difference by Chris Voss. Highly recommend reading it :)
Repeat the last exact words that the participant said and pause… This works like magic.
People don’t like silence…they want to fill it. They will try to explain what they just said, and you will be surprised how it is different from the first answer they gave you.
- At the end of the interview, ask, "Is there anything else you want to share with us?". At the end of the interview, people feel more comfortable sharing what is important to them. So leave at least 5 min for this question.
- Thank your participant and say how valuable for you this interview was. Then, ask if you can reach out to them again for interviews, and if the answer is yes — put them into your database, so you don't lose contact.
Analyse user interviews
When the round of interviews is finished, it's time to do a research analysis. I do it in Miro and will share some screenshots with you.
Step 1. Collect all the notes from user interviews and put them as stickers.
Put tags on each sticker:
- Month and year of the interview (Mar 2023)
- Persona (Pro user)
- Participant number (P1)
Step 2. Create a board divided into sections:
- Gains
- Jobs/Tasks
- Pains
Afterward, group interview notes by topics and put them into these sections.
If a couple of participants shared the same thought — combine them in one sticker and add the participant numbers. It also can happen that two or three personas share the same thought. You can put a couple of personas on the same sticker as well. Your sticker can look like this:
Share research insights with stakeholders
It's your choice in what format you present research insights to stakeholders. It can be a presentation, spreadsheet, Miro board, etc. More importantly, do it as soon as you finish the research analysis, so the product and design team use up-to-date insights.
Use the hypotheses you created for the Research Plan and connect them to the interview notes. This approach makes it easier to see if you confirmed or disproved your hypotheses.
On this template, you can see that hypotheses are connected to the notes. Change the color of hypotheses depending on results (red — if the hypothesis is disproved and green — if confirmed).
The yellow section is for "Other insights." Place here everything that does not match your created hypotheses but is interesting for further exploration.
After you finished the research analysis, your summary can look like this:
Thank you for reading! Share your thoughts & questions in the comments, and I will be happy to answer them :)
