Vendor case studies, including the ones we publish about company headshot rollouts, have a known defect: they are selected. A published one is, by construction, one that went well enough to publish. You are reading the top of a distribution and the shape of the rest is invisible.
That is true of every vendor's page, ours included. It does not make them useless, it makes them useful for a different thing than they are presented for.
Read for constraints, not outcomes
Outcomes are chosen by the person writing. Constraints leak, because they have to appear for the story to make sense.
A constraint tells you which situations the product suits. That is a lower claim than "this works" and a far more useful one when you are the person who has to integrate it.
Ours are a decent worked example. The Wyn Solutions write-up has a constraint buried in it that matters more than the result: the team was distributed, so getting everyone in a room with a photographer was never available. If your team sits in one building with a photographer on retainer, that story is not about you, and its success predicts nothing about your case.
The BetterHealth Group one leaks a different constraint: healthcare compliance plus meaningful staff turnover. A one-off session ages badly when a fifth of the faces change annually, which changes the maths from "a project" to "a recurring process". Again: does that resemble you?
The Elucient write-up is the smallest of the three, and small is its own signal about where the product fits comfortably.
Three questions to ask of any of them
What was the alternative? "Better than nothing" and "better than the agency we already paid" are wildly different findings, and write-ups are usually vague here on purpose.
What is the denominator? A 40% improvement on 10 people is 4 people. Percentages without a base rate are decoration.
Who is quoted? The person who ran the project knows what went wrong. The executive who approved the budget knows what it cost. A quote from the second tells you much less.
The version for engineers specifically
Case studies almost never mention integration. If you are the one wiring it up, the questions you need answered are not in there:
- What is the API surface, and is it versioned?
- What happens on partial failure in a batch of 200?
- Is there a webhook, or are you polling?
- What is the rate limit, and does it apply per account or per key?
- Can you export everything if you leave?
None of that appears in a customer story, and all of it determines whether the project is a week or a quarter. Ask separately.
Where ours live
All three are collected on the case studies page, and the healthcare one is also published on the main site with slightly different framing, which is itself a small lesson in how these get tuned per audience.
Read the constraints. If none of the three resemble your situation, that is genuinely useful information, and it is information the page was not trying to give you.
More from this series
- Test any generation tool on the hardest face you can find, not the easiest
- Crop ratios and framing conventions across 5 professional photo genres, as a reference table
- Colour management for portfolio sites, and the 4 professions where getting it wrong loses the client
- La photo sur le CV en France : usage tenace, obligation nulle, et l'ATS n'y est pour rien
On BetterPic
Top comments (0)