Case studies

The case study most companies publish names an outcome. The one worth reading names the alternative the customer rejected — because that's the sentence a prospect is actually looking for. They already believe a good outcome is possible. What they don't know is whether anyone like them chose this over the thing they're currently leaning toward.

The operational shape

Trigger. A deal marked won, plus a customer approval flag. Both. A won deal without the flag isn't a trigger, it's a request to go ask.

Reads. Customer stories from entities, positioning and cleared proof from decisions.

Cadence. Per closed-won deal, which means the supply is set by the sales motion and not by the content plan.

Gate. The strictest in the system: approve before publish, plus customer sign-off. Named customers and any statistic both sit behind a named human, and here they arrive together.

What good looks like

It names the alternative. What the customer was going to do instead, and why they didn't. This is the whole page. Everything else is supporting evidence.

The numbers are ones the customer will defend. Not the most flattering available — the ones their own team would repeat in a meeting where someone might push back. A number the customer would hedge on in person is a number that will be hedged on when a prospect calls them for a reference, and that call is the point of having the case study at all.

It admits what was hard. A case study where nothing went wrong reads as marketing, because nothing goes entirely right. The implementation detail that cost two extra weeks is the detail a buyer trusts, and it costs nothing to include once the customer has approved it.

The customer sounds like themselves. Quotes assembled from what someone would plausibly have said are the fastest way to make a real customer look invented. If the interview didn't produce a usable quote, publish without one.

When it goes stale

Case studies decay in a way that's specific to them: they stay technically true and stop being representative.

Four triggers. The named person leaves — a quote attributed to someone who no longer works there is checkable, and a prospect will check. The customer's situation changed enough that the story now describes a company they no longer are. The number aged past the window where it means what it meant. And the product moved past it, so the story describes a workflow the current product doesn't have.

The last one is the most damaging, because it makes the newest buyer's experience contradict the page that sold them.

What it feeds

• Battlecards — the rejected-alternative section is the raw material
• Value propositions — a claim with a customer behind it outranks a claim without one
• Comparison pages — the only place a competitor comparison can cite a real decision
• Sales decks and proposals — instantiated per account, from the same source

Case studies are the highest-leverage capability in the convince category and the slowest to supply, which is why the ones that exist get reused rather than replaced.