UX research for beginners: how to run your first study

Step 1

Write down what exactly you want to learn

Research without a question turns into a nice chat with no result. So first write a research question: one or two sentences about what you don't know that affects the design. For a study project, pick a familiar situation your friends deal with: booking a doctor, renting an apartment, choosing a course, ordering food.

• •weak: "Do people like food delivery apps?" — the answer tells a designer nothing;
• •better: "How do people choose where to order dinner, and what makes them abandon an order?";
• •also write down which decisions you'll make from the results: for example, what to show on the first screen.

💡 If you can't put the question to a single real person, it's still too abstract. Narrow it until you can picture who you'll ask.

Step 2

Pick a method that fits the question

Nielsen Norman Group maps UX research methods along three dimensions. The first is what people say versus what they do: surveys and interviews reveal opinions and beliefs, while observation and usability tests reveal behavior, and the two often differ. The second is qualitative versus quantitative: qualitative methods explain "why" and "how", quantitative ones answer "how many" and "how often". The third is context: whether people use the product as they normally would, follow the researcher's script, or don't use the product at all.

• •to understand needs, habits and pain points: interviews;
• •to check whether your mockup can be used: usability testing;
• •to collect open answers from more people: a qualitative survey with open questions;
• •to see how people group information (menus, categories): card sorting.

💡 Interviews suit a first study best: they don't need a finished design, and they teach the key skill of listening without leading.

Step 3

Find participants and prepare an interview guide

Participants should have real experience of your topic: if you're researching apartment rentals, talk to people who recently looked for an apartment, not to anyone at all. For a study project, start with a few people from your circle or from topic communities, but avoid those who know your idea and want to please you. Before the conversation, explain why you're doing it and how you'll use the answers, and ask permission to record. According to the GOV.UK Service Manual, you must get informed consent from every participant, and they can withdraw it at any time.

• •write an interview guide: an introduction, a few easy warm-up questions about context, the main topics, a wrap-up;
• •ask about specific past experience: "Tell me about the last time you looked for an apartment";
• •avoid leading questions: not "That was inconvenient, wasn't it?" but "How did it go?";
• •don't ask people what a feature should be: they describe their problems better than solutions.

💡 Run the guide with a friend before the first real interview. One practice conversation shows which questions are unclear.

Step 4

Run the interviews and listen more than you talk

In an interview, your job is to let the other person talk. Ask an open question, hold the pause and follow up: "Why?", "What happened next?", "Can you give an example?". Don't correct the participant or explain the "right" way. If you can, invite someone to take notes so you can focus on the conversation.

• •write down verbatim quotes: you'll need them for findings and for the case study;
• •note actions and workarounds: "I write it in my notes", "I ask in a chat", "I call my mom";
• •record emotions and moments where the person stops or hesitates;
• •right after each conversation, write down the key observations while your memory is fresh.

Step 5

Analyze the data and draw conclusions

Move your observations onto separate notes: one idea per note. Then group similar ones into clusters. This technique is called an affinity diagram, and it's easy to do on a wall with sticky notes or on any online whiteboard. Name each group as a finding, not a topic: not "Price" but "People don't trust the price until they see delivery costs".

• •keep what you saw or heard separate from your own assumptions;
• •keep quotes as evidence for each finding;
• •write a few key insights and, for each, what changes in the design;
• •be honest about limits: how many people, who exactly, what the study doesn't cover.

💡 Show your findings in the community and ask: "Does this insight follow from the quotes?" Fresh eyes quickly catch findings you "filled in" yourself.