Key takeaways
- You do not need a large research team to learn from customers.
- Start with the decision the research must support.
- Combine conversations, task testing and existing data.
- Avoid leading questions and over-generalising from small samples.
In this article
Start with the decision research must support
Research is most useful when it informs a specific decision. Before choosing methods, write down:
- What decision are we trying to make?
- What do we already know, and what are we assuming?
- What would change our plans?
For example: "Should our new website prioritise online booking or quote requests?"
Interviews and customer conversations
Interviews help you understand motivations, language and context.
- Talk to recent customers, lost leads and people who match your target audience.
- Ask about past behaviour rather than hypothetical future choices.
- Use open questions such as "Tell me about the last time you..."
- Record with consent, or take careful notes.
Usability testing with key tasks
Ask participants to complete realistic tasks on your website, a competitor's site or a prototype while thinking aloud. Watch where they hesitate or make mistakes. Even a few sessions often uncover significant issues. Testing a prototype before development is especially cost-effective.
Surveys and behavioural data
| Method | Best for | Watch out for |
|---|---|---|
| Short surveys | Measuring preferences across many people | Leading or confusing questions |
| Analytics | Seeing what people actually do | Data without the reasons behind it |
| Site search terms | Discovering what people look for | Small volumes on smaller sites |
| Enquiry and support records | Common questions and pain points | Bias towards problems |
Turning findings into priorities
Group observations into themes, note how often and how severely each issue appeared, and link each recommendation to evidence. Share findings visually with quotes and short clips so the whole team understands them.
Avoiding biased questions and overgeneralisation
- Avoid questions that suggest the answer, such as "Don't you think this is easier?"
- Do not treat a single opinion as a universal truth.
- Separate what people say from what they do.
- Be clear about who participated and who did not.
See our user research and UX / UI design services, read about UX audits, or contact us.
Frequently asked questions
How many people should we interview?
For exploratory interviews and usability testing, a small number of well-chosen participants often reveals the main patterns. Larger samples are needed for surveys and statistical confidence.
Can we interview our own customers?
Yes. Existing customers are a valuable source of insight. Explain the purpose, keep sessions short and respect their time and privacy.
Should we pay participants?
Offering a modest thank-you, such as a voucher, is common and improves participation. Make sure it does not bias responses.
Can you run research for us?
Yes. We plan, recruit, run and analyse research and turn the findings into practical recommendations.
Related services
Melbourne Design Studio is a design and development studio in Truganina, Melbourne, with more than 20 years of design and user experience work. Our team plans, designs and builds websites, apps, brands and campaigns, and writes these guides to help business owners make informed decisions.
Meet the team
