05 Nov Listening Like a Designer: Empathy Interviews and Research a Small Business Can Actually Do
*Part 2 of Start the UX Design Process — a series on the UX principles I use in client work.*
Ask a business owner what customers think of their website and you usually get “everybody loves it” — from the owner, the staff, and the family — or a shrug. Neither is research. Both are guesses, and guesses are what the first post warned against: designing for people in your imagination instead of people in real life.
The fix is not complicated, and it does not require a research department. It requires listening on purpose. In UX work that kind of listening has a name: empathy interviews. Strip away the name and it is simple — a short, open conversation with a real user, where your job is to understand their world, not to sell them yours.
### What an empathy interview is — and what it is not
An empathy interview is not a sales call, a customer satisfaction survey, or a chance to show off the new homepage. You are not asking “do you like our website?” because that question tells you almost nothing. People are polite. They will say yes, and then they will go back to struggling with the same page.
A good empathy interview asks about the person’s job, not your website. What were you trying to get done the last time you visited a site like ours? Walk me through what happened. Where did you get stuck? What did you do instead — call, give up, go somewhere else? The answers that matter most usually arrive in the second or third sentence, after the polite answer runs out.
That is why the best interview questions are open and boring-sounding. “Tell me about the last time you…” beats “did you find our site easy to use?” every time. One invites a story. The other invites a grade. Stories are where the design problems hide.
### Before you ask anyone anything: set the goal
The most common interview mistake is writing the questions first. Write the goal first instead. A research goal is one plain sentence about what you need to learn, tied to the problem your site is trying to solve — for example, “I want to understand what frustrates first-time visitors who try to register on their phones.” That sentence does three jobs: it decides who you should talk to, it decides which questions are worth asking, and it decides later whether an answer actually matters.
Who you talk to follows from the goal. You want a mix that looks like your real audience, not your easiest audience — different ages and familiarity levels, phone-first people, and people with disabilities, who surface problems everyone else quietly works around. A manageable group that resembles your visitors beats a large group of the same kind of person.
Then write the questions, and test each one: relevant to the goal, answerable with more than yes or no, short and plain, not steering the person toward an answer you want, and open to a follow-up like “tell me more about that.” Consistent questions across interviews let you compare answers honestly, while still leaving room for a real conversation.
### Imagine the interview before you run it
There is one more step before any real conversation, and it is the step most small projects skip: imagine the interview first. Pick a few short, made-up sketches of the kinds of visitors you expect — a busy parent, a first-time donor, a guest who has never been to your church — and walk your own questions through their eyes. How might each of them answer? Where might a question confuse them, or make them clam up?
That dry run catches leading wording before it costs you a real interview: a question that assumes everybody has struggled will make people who have not struggled invent one, so reword it neutrally and save the sharper follow-up for when the visitor raises the topic themselves. It also reminds you that these sketches are not personas — just warm-up guesses to sharpen your questions. Personas come later, after real research.
### Five conversations beat fifty opinions
Small businesses sometimes freeze at research because they imagine they need hundreds of responses. You do not. For a typical small business or nonprofit website, five honest conversations with the right people will surface the big problems — the same stuck points start repeating by the fourth or fifth, and that repetition is the signal.
Who counts as “the right people”? Not your staff, not your friends, and not your most loyal customer who would forgive anything — the mix from the goal section: somebody new, somebody who almost did not convert, somebody phone-first, somebody who found you through a search. For a church site, that means a first-time guest, not just members. For NC Freedom Fest, the parent in the car on event day, not just the committee. For United Way of Wayne County, the first-time donor or volunteer, not the staff who built the page.
### The interview habits that actually work
Keep it short — twenty minutes is plenty. Before you start, treat the questions as a guide, not a script: know them well enough to follow a surprising answer, bring what you need, learn what you can about the person first, and run through the questions once out loud so you find the clumsy ones before a real visitor does.
When you meet, set the tone first: easy conversation, a thank-you for their time, and the basics that matter for the goal, before the real questions begin. People tell the truth when they feel at ease and valued, and they clam up when they feel graded.
Then mostly listen. Ask about the past, not the future — “what did you do last time?” gives behavior, while “what would you like us to add?” gives a wish list. Follow the stuck points: when somebody says “I couldn’t find it,” ask what they clicked, what they expected, and what they did next. Write down exact words, not summaries, plus what you observed — mood, hesitation, anything that might have skewed an answer. If you record, ask permission first and use it to check your notes, not replace them. At the end, leave room for final thoughts and thank the person again; how they leave is part of how they will remember your organization.
Keep that exact-words habit in mind: if a visitor says “I didn’t know where the kids went,” that phrase is worth more than a paragraph of your interpretation, because it is the language your site should answer in.
And one habit to avoid: defending the site mid-interview. The moment you explain why a page is the way it is, the person stops telling you the truth and starts agreeing with you. Save your explanations for the design phase. The interview belongs to them.
### What you do with what you hear
Listening is only half the job. The other half is refusing to file it away. After five conversations, sit down and turn your notes into empathy maps — one per person, then a combined map for the group — and the stuck points that repeated, in the visitors’ own words, will stand out on their own. Those maps are the next post in this series, and they become the raw material for the personas in post four, and eventually the one clear problem statement in post seven. Research that never changes a design decision was not research — it was a nice chat.
So start small. Pick five people who actually use — or tried to use — your site, and ask them about the last time they tried. You will learn more in a week of those conversations than in a year of guessing.
—
*Want help turning what your visitors say into a site that works? I build and redesign websites for small businesses, non-profits, and churches across Eastern NC. [Get a free quote](/) or [see recent work](/web-design/).*
*Previous in this series: [Why Good Websites Start With Users, Not Layouts](/why-good-websites-start-with-users/) · Next: [Finding Real Pain Points: What’s Actually Frustrating Your Website Visitors](/finding-website-visitor-pain-points/)*
> *This article is my own take, in my own words, on UX principles I use in client work. It is not affiliated with or endorsed by Google or Coursera.*