Case study: writing one a buyer actually believes

A case study answers one question for a buyer: has this worked for somebody like me. That is a narrower job than most case studies attempt, and it explains why the typical version — glowing, vague, structurally identical to every competitor's — persuades nobody. Credibility comes from specificity, and specificity is exactly what gets removed during approval.

What makes a case study credible

• A named customer, or at minimum a described one: sector, size and situation, since «a leading provider» describes nothing.
• The situation before, including what they had tried that did not work.
• What was actually done, in enough detail that a reader can picture it rather than admire it.
• Numbers with a baseline — a figure without a before is a claim, not a result.
• A timeframe, because «improved efficiency» over six years is a different story from over six weeks.
• Something that did not go perfectly, which does more for believability than any superlative.

The numbers that matter

Use the customer's metric rather than yours. They did not want more logins or better adoption; they wanted to stop something that was costing them. The strongest figures are the ones they were already tracking before you arrived, because those are the ones they trust and the ones a similar reader will recognise. Where no number exists, a specific qualitative statement beats an invented percentage — «the monthly close went from four days to one, and the finance manager stopped working weekends» is checkable, memorable and does not require a measurement nobody made.

Get approval for the quote and the figures separately and in writing, and ask at the same time whether the customer's name may be used. The commonest failure is a finished case study that cannot be published because the person who agreed informally has left and nobody can produce their permission.

Getting it written without losing the interesting parts

1. Interview rather than send questions, because the useful detail arrives in asides rather than in answers.
2. Ask what nearly stopped them buying, which is the most valuable paragraph you will get.
3. Write it, then send it for approval with the specifics intact and let them remove things, rather than pre-sanitising.
4. Offer to anonymise the company but keep the numbers, which is often an acceptable trade when the first answer is no.
5. Date it and keep it accurate, since a five-year-old study describing a product that has changed undermines the rest.
6. Make it easy to find: one page per study, linked from the relevant product page rather than buried in a resources index.

Ettex Web suits a case study as its own page, so it can be linked directly from a proposal and found in search rather than living inside a brochure. The conversation that produced it belongs in the CRM, because the questions that nearly stopped that buyer are the ones the next one will ask.