Foldwren

User research on a small budget

How a small business can do useful user research by talking to five customers, asking better questions and watching how people actually use things.

You do not need a research department to find out what your customers want from a website or an app. You need a few conversations, a notebook, and the discipline to listen more than you talk.

Why bother before you build?

Because the most expensive thing in any design project is building the wrong thing well.

Owners often carry a confident picture of what customers want, and it is usually partly right. The gap is in the details. An accountant assumes clients want a portal to view documents, when what they really want is to stop being asked for the same paperwork twice. A cafe owner plans a loyalty app, when regulars mostly want to skip the queue.

A handful of conversations will not give you certainty. They will catch the big misunderstandings, and those are the ones that waste money.

Who should you talk to?

Five people is a good start. Not five friends, and not your five happiest customers.

Aim for a mix:

  • A couple of regulars who know you well.
  • One or two newer customers who still remember what it was like to find you.
  • Someone who enquired and did not go ahead, if you can get them.
  • If the tool is for staff, the people who will use it daily, not only their manager.

Ask in person or by a short message: "We are improving how people book with us and I would value twenty minutes of your honest opinion." Most people are flattered to be asked. A coffee or a small discount is a fair thank you.

What should you ask?

The golden rule: ask about what people have done, not what they think they would do. People are poor at predicting their own future behaviour and very good at describing last Tuesday.

Questions that work:

  1. "Tell me about the last time you booked with us. How did you go about it?"
  2. "What were you doing just before that? What prompted it?"
  3. "Was there anything that slowed you down or annoyed you?"
  4. "Have you ever started to book or order and given up? What happened?"
  5. "How do you handle this with other businesses like ours?"
  6. "If you could change one thing about dealing with us, what would it be?"

Questions to avoid:

  • "Would you use an app if we had one?" Almost everyone says yes to be polite. It tells you nothing.
  • "Do you like this design?" Opinions about looks are cheap and unreliable.
  • "Would you pay for this?" Hypothetical money is not real money.

Then be quiet. When someone pauses, resist the urge to fill the silence. The useful detail usually comes in the second half of an answer.

Watching beats asking

What people say and what people do are different things, and when they disagree, believe the doing.

If you have an existing website or booking process, ask a customer to use it in front of you. Give them a task, such as "book a haircut for next Thursday afternoon", hand over a phone, and watch without helping. You will see hesitations, wrong taps and workarounds that nobody would ever think to mention in an interview.

You can also observe without setting anything up:

  • Stand near the counter and note the questions customers ask most.
  • Read back through a month of enquiry emails and group them by theme.
  • Ask your front of house staff what they explain over and over.
  • Look at which pages on your site people visit before they contact you, if you have Google Analytics or something similar installed.

Turning notes into decisions

Raw notes are not much use until you sort them. Straight after each conversation, write down the three things that stood out. Once you have done all five, lay the notes side by side and look for repeats.

A simple way to sort:

  • Heard from three or more people: treat it as a real pattern and act on it.
  • Heard once: interesting, but do not redesign around one person.

Then turn each pattern into a plain statement. "New customers cannot tell which service they need." "Regulars want to rebook the same thing without filling in the form again." Those statements become the brief for whoever designs the next version, and they are far more useful than "make it modern".

Be wary of treating feature requests as instructions. When someone says "you should have an app", ask why. The reason underneath, such as "I can never find your booking link", is the thing to solve. There may be a much cheaper fix.

A plan you can run in a fortnight

  1. Write one sentence describing what you want to learn.
  2. List five people and invite them.
  3. Prepare six open questions about past behaviour.
  4. Hold the conversations, twenty to thirty minutes each. Take notes or record with permission.
  5. Watch at least two of them complete a real task on your current site.
  6. Sort what you heard into patterns.
  7. Pick the top three problems and bring them to your designer.

That is perhaps four or five hours of your time. Set against the cost of building something nobody asked for, it is the cheapest part of the whole project.

Your order