What five conversations can teach you before a redesign

Most website redesigns start with opinions. The board thinks the site looks dated. A staff member wants the events page on the home page. Someone has seen a competitor’s site and liked it. All of those opinions might be right, but none of them come from the people the website is actually for.

Before we redesign anything, we try to talk to a few of those people. Not a survey of hundreds, not a formal study: just a handful of conversations. Five is a good number to aim for. It’s small enough to fit into a busy fortnight, and big enough that patterns start to show.

Why so few is enough

Small-scale research isn’t trying to measure anything. It won’t tell you what percentage of visitors want a feature. What it does well is reveal things you didn’t know to ask about: the word people use that isn’t on your site, the step where they always give up, the question they phone you about because the website never answers it.

By the third or fourth conversation, you’ll often hear the same thing again. That repetition is the signal. When two or three people from different backgrounds independently stumble on the same page, you’ve found something worth fixing.

Who to talk to

Pick people who use, or should use, your website, and who aren’t on your staff or board. Insiders know too much; they can’t un-know where things are.

Good places to find people:

• Recent customers, clients or members who joined in the last year
• People who enquired but didn’t go ahead
• Parents, volunteers or supporters, depending on who your site serves
• Someone who fits your audience but has never heard of you

Try to include at least one person who finds technology harder, or who uses a screen reader, magnification or a phone on a slow connection. They’ll show you problems everyone else is quietly working around. A small thank-you, such as a voucher or a coffee, is a fair way to respect their time.

What to ask

The aim is to hear how people think, not to collect compliments. Ask about what they actually did, not what they might do.

Start with their world

Ask about the last time they needed what you offer. How did they look for it? Who else did they consider? What nearly stopped them? This gives you context before they ever see your site.

Then watch them use the site

Give them a realistic task, like “find out whether this program suits your child” or “work out how to book”, and ask them to talk out loud as they go. Then stay quiet. The hardest part of this research is not helping. When they pause or go the wrong way, that’s the most useful moment of the session.

Questions that work

• “Tell me about the last time you…”
• “What were you expecting to find there?”
• “What would you do next?”
• “Was there anything that made you hesitate?”

Avoid questions that suggest an answer, such as “Did you find the menu easy?” Most people are polite, and they’ll tell you what they think you want to hear.

Turning conversations into decisions

Take notes during each session, or record it with permission, and write a short summary straight afterwards while it’s fresh. Handle what you collect with care: tell people how their information will be used, keep recordings only as long as you need them, and leave names out of anything you share.

Once you’ve done your five, put the summaries side by side and look for things that came up more than once. We usually sort them into three groups:

• Things to fix: places people got lost, confused or stuck
• Things to keep: what worked well and shouldn’t be lost in a redesign
• Questions to explore: hunches that need more evidence before acting on them

That list becomes the brief. It’s far more useful than a mood board, because every item traces back to something a real person did or said.

Our rule of thumb

If you can’t name a real person who struggled with something, be careful about redesigning it. And if three people struggled with the same thing, fix it before you change anything else.

Where to start

You don’t have to wait for a redesign. Pick one important task on your site, find three people who might do it, and watch them try. You’ll learn something useful in an afternoon. If the conversations point to a new tool or feature rather than a fix, it can be worth testing the idea cheaply first, as we described in prototyping before you build.

If you’d like help planning or running research before a redesign, see our research service, or tell us what you’re working on. We’re happy to help you work out what to ask and who to ask.