Case Studies That Close
The strongest evidence a practice can publish: a named client, a real before-and-after, numbers a sceptic can check. The anatomy, the consent conversation, and the mistakes that neuter them.
At the top of the fakeability hierarchy, above even the review record, sits a form of proof most practices never build: the case study. A real client, named with permission, a real situation, what was done, and what measurably changed. It outranks everything else for one reason — it's checkable. A sceptical prospect could, in principle, verify it. That possibility is what they're buying when they believe you.
One genuine case study converts harder than a testimonial wall, because it answers the question every nervous buyer is actually asking: not "are you good?" but "have you done my exact problem before, and how did it go?"
The anatomy of a case study that closes: a named client (or a precisely-described one, where naming isn't possible); the situation in the client's terms, stakes included; what was actually done, plainly; the outcome as numbers a sceptic could check; and one honest wrinkle — the thing that was harder than expected. The wrinkle isn't a confession, it's the credibility engine: flawless stories read as marketing; specific difficulty reads as memory.
Why numbers, and which
"We helped them grow significantly" is a testimonial in a suit. "Days-to-paid fell from 54 to 11" is evidence. The numbers don't need to be dramatic — they need to be particular: the before, the after, the timeframe. Particularity is what a reader's scepticism can't dismiss, and what makes the study quotable by assistants besides.
Where results genuinely resist measurement (much legal and medical work), measure the experience instead: timelines met, steps avoided, what the client said after — in their words, attributed.
The consent conversation
The reason practices don't publish case studies isn't writing — it's the imagined awkwardness of asking. In reality, the ask lands well with the right client at the right moment (the same delight moment that produces reviews): most people are glad to have their problem's solving told, provided they control the telling.
Make consent easy to give: they see the draft before anything publishes; they choose the naming level (full name, first name, "a dental practice in Bedford"); they can withdraw later. For regulated professions, anonymised-but-particular is fully workable — precision of detail substitutes for the name. What can't be workable is invented composites presented as real; one discovered fabrication poisons every true thing on your site.
The mistakes that neuter them
Writing it as a brochure. The study's voice should be a professional recounting a job, not a marketer celebrating one. Flat, specific, slightly understated — the register of someone with nothing to prove.
Burying it. A case study nobody encounters converts nobody. Link it from the service page it proves, from proposals, from anywhere a prospect hesitates.
Making it about you. The client is the protagonist; you're the competent help. Studies that read as self-portraits persuade no one — the reader is casting themselves in the client's role, so give them a client to identify with.
Stopping at one. One study proves an anecdote; three spanning your main services prove a pattern. Past five or six, add only when a study covers a new situation.
Questions practices actually ask
How long should one be? Five hundred words is usually plenty — situation, work, numbers, wrinkle, quote. Long enough to be particular, short enough to be read at 9pm.
Client won't be named. Worth publishing anyway? Yes, if the details stay sharp: "a three-partner law firm in Dartmouth" with real numbers beats a named client with vague ones. Specificity is the active ingredient; the name is a multiplier.
Case study or testimonial? Different jobs: the testimonial says believe them; the case study says here's what happens when they do your problem. Testimonials decorate; case studies close. Use quotes inside studies and get both.
Where do the numbers come from if I never measured? From the next engagement: note the before-state at the start of every substantial job from now on. A measurement habit at intake is a case-study pipeline — and it's the same instrument panel your own practice should be running anyway.
Part of Earn Trust — the evidence layer.