User research interviews produce a strange kind of praise. You sit a customer down to understand how they work, promise them it's not a sales call, and somewhere in the middle they say the truest thing anyone's ever said about your product — not because you asked them to endorse it, but because they were describing their day and your product happened to be the good part of it. That offhand honesty is exactly what makes it worth quoting and exactly what makes it dangerous to touch. A research interview runs on a promise: this is to help us improve, not to market to you. The moment you lift a line onto your homepage, you're using something said under that promise for a purpose the person didn't sign up for. It can be done well, but only if you treat the transcript as borrowed, not owned.
Why a research quote is unusually strong — and unusually fragile
The reason a line from a research call lands harder than a solicited testimonial is that nobody was performing. When you ask a customer for a testimonial, some part of them writes for your benefit; in a research interview, they're answering your questions about their own workflow, and the praise slips out sideways. "Honestly the reporting is the one thing I don't have to think about anymore" is more convincing than any sentence a customer would compose when they know it's going on a website, because it wasn't composed at all.
That same quality is what makes it fragile. The strength comes entirely from the context — an unguarded answer in a private, low-stakes conversation. Strip that context away and publish it as a marketing quote and you've done two things at once: taken the person's words out of the frame they meant them in, and quietly changed the deal you offered when you set up the call. The customer agreed to help you learn. They did not agree to become a case study.
The consent problem is the whole problem
Everything about doing this well comes down to one fact: research consent and marketing consent are different consents, and you only have the first one. A recording release you signed at the start of an interview almost always covers internal use — analysis, notes, sharing the transcript with your team. It rarely covers publishing a quote with the customer's name attached on a public page. Assuming the first implies the second is the single most common way teams turn a goodwill interview into a customer who feels used.
So the rule is simple and non-negotiable: you go back and ask, specifically, for the thing you now want to do.
- Quote the exact line back to them. "During our research call you said X — we'd love to feature that as a testimonial on our site. Would you be comfortable with that?" This is a different question from anything they answered during the interview, and it deserves its own yes.
- Say where it will appear and how they'll be named. Homepage with full name and company is a bigger ask than a first-name-and-role quote in a blog post. Let them choose the attribution level, the same way you'd let them for any testimonial whose specific claim you need to be able to stand behind.
- Give them the transcript context, not just the sentence. Showing the surrounding exchange reminds them what they meant, and lets them catch a line that reads worse out of context than it felt in the room.
If the answer is no, or no reply, the line stays in your research notes. A great quote you can't use is not a loss — it's a customer relationship you didn't spend.
How to lift the line without distorting it
Once you have consent, the discipline is preservation. The whole value is that they weren't performing, so anything you do to make the line "sound like a testimonial" destroys the reason it's worth having.
- Keep the exact wording, including the hedges. "It's basically the one thing I don't have to babysit" should not become "an indispensable part of our workflow." The casualness is the evidence. Smoothing it into marketing English is the same mistake as flattening any thank-you into a polished claim it never made.
- Don't stitch two answers together. If the customer said one good thing at minute three and another at minute twenty, you may quote either, but you may not splice them into a single sentence they never spoke. Provenance is the point; a manufactured composite has none.
- Cite the source honestly, without over-claiming. "Shared with permission, from a customer interview" is accurate and quietly powerful — it signals the person said this while describing their own work, not while doing you a favor. You don't need to dress it up as a spontaneous rave; the plain provenance is the persuasion.
The lines you cannot cross
You cannot publish a research quote on the strength of a research release alone — a new use needs a new yes. You cannot attach a name the customer didn't agree to see attached, even if the quote is glowing. You cannot edit the sentence to say more than they said, or splice their answers into praise they never delivered in one breath. And you cannot lean on the "it's not a sales call" framing to get candor and then route that candor straight to your marketing team as if the promise never happened. Every one of these trades the quote's only real advantage — that it was said honestly, off-guard, to help rather than to sell — for a short-lived polish that collapses the instant the customer sees their unguarded words repurposed as a pitch.
The mindset
A research interview is one of the few times a customer tells you the truth about your product with nothing to gain. That honesty is worth more than any solicited testimonial, which is exactly why it deserves more care, not less. Treat the transcript as borrowed: ask again for the specific use, keep the words exactly as spoken, name the source plainly, and let a customer who said something true in private decide whether it's said in public. Do that and a line you overheard while trying to learn becomes the most credible proof you own — a real person, describing their real work, who happened to tell you the truth and then agreed to let you repeat it.