a case study that wins work.

most case studies are written from the inside out: here is the client, here is what we produced, here are some screenshots, here is a compliment. the reader learns what you made and nothing about whether you can solve their problem.

the case studies that generate enquiries are structured the other way round — the problem first, described precisely enough that a prospect with the same problem stops reading casually and starts reading carefully.

the shift is uncomfortable because it means writing mostly about the client and barely about yourself. that is also why it works: a prospect is not evaluating your creativity, they are checking whether you understand a situation they are currently in.

— 01the problem, in the client's words.

open with the situation as it was: what was happening, what it was costing, what had already been tried, what the constraint was. specificity is what makes this work — "leads had dropped" is generic, "enquiries were steady but the sales team was spending three days a week on people who could not afford the service" is recognisable.

the person you want reading this is looking for evidence that you have seen their problem before. every sentence of accurate diagnosis is worth more than a paragraph about your process.

and include the constraint honestly. a small budget, a short timeline, a regulated category, one person doing the marketing, no existing content. constraints make the result credible and they make you relevant to the many prospects who have them. a flawless result achieved with unlimited budget is impressive and useless to the reader who has neither.

this section should be long enough to feel uncomfortable. most agencies spend two sentences here and six paragraphs on their own work, which is exactly backwards. the easiest way to get this right is to interview the client about the period before you arrived and quote them, rather than writing the situation from memory.

— 02the decision, not the deliverables.

the middle of a case study should explain what you decided and why, not list what you produced. a list of deliverables proves you can invoice; a decision explains how you think, which is the thing being evaluated.

so name the choice and the alternative you rejected. "we moved the budget out of awareness video and into search, because the category had existing demand and the client needed enquiries within the quarter" tells a reader something. "we developed a full-funnel content strategy" does not. the rejected alternative is the part that proves a decision was made rather than a package sold.

one clear decision beats five. a case study with a single legible pivot is remembered; one that lists eleven workstreams reads as an invoice and is forgotten before the page ends. if several decisions genuinely mattered, write several case studies rather than one crowded page.

— 03a number, with its starting point.

a result without a baseline is not a result. "three hundred per cent increase" from an unstated base could be three enquiries to twelve, and experienced buyers assume the least impressive interpretation when the starting point is missing.

so give the before and the after, the timeframe, and the metric that mattered commercially rather than the most flattering one available. enquiries, cost per acquisition, revenue, booked appointments, retention — not reach or impressions, which read as evasion in a case study. if the only available number is a reach figure, the honest move is to describe the outcome qualitatively rather than dress up a soft metric as a result.

if the numbers are confidential, use ratios or ranges and say that is what you are doing. an honest "cost per lead fell by roughly a third over four months" is more credible than a precise number nobody can verify, and far better than a vague claim of success. get sign-off on the exact wording before publishing, since a client who feels exposed by a number will withdraw permission for the whole piece.

— 04what to leave in that everyone removes.

include what did not work. a case study that mentions the format that failed before the one that worked is dramatically more believable than a clean narrative, and it signals that you test rather than guess.

include the client's contribution too — the access they gave, the decisions they made quickly, the material they provided. it is accurate, it flatters them, and it quietly tells the next prospect what the relationship requires of them. that last effect is underrated — a case study that shows what good clients do attracts more of them.

and keep it short. one page, one problem, one decision, one number, one honest complication. we write ours that way and link them from the service pages rather than in a gallery, because a case study is most useful next to the service it proves — the same reason the reporting is built to produce those numbers in the first place.