A picture cannot argue

A picture does not argue

Put an image of a finished poster in front of a buyer and you have told them one thing: you can make a poster that looks like that. You have told them nothing about whether you understood the problem, whether the client got what they needed, or whether you would be good to work with. Those three questions are what they are actually trying to answer, and an image cannot answer any of them.

A case study answers them in about 200 words. That is the whole technique. Not a design blog essay, not a process documentary: roughly four short paragraphs sitting under the images.

The four parts

The situation, in the client’s words. One or two sentences, concrete and dated. “A fourteen-year-old tuition centre in Indore was losing enquiries to a newer competitor two streets away whose signage looked more current.” Not “the client needed a brand refresh”. The specificity is doing two jobs: it proves the project was real, and it lets a reader with a similar problem recognise themselves.

The constraint. Every real project has one, and naming it is the single strongest signal of experience in the whole document. Budget under a certain number. Three weeks because the new term started. Had to be printable by a local vendor with two spot colours and no lamination. Had to work in Hindi and English at the same size. Amateurs hide constraints because they feel like excuses. Buyers read them as evidence you have worked inside reality.

One decision and why. Not every decision: one. The one that a different designer would plausibly have made differently. “I kept the original 1998 wordmark and rebuilt everything around it, because the families who refer this centre have been walking past that word for a decade and the name recognition was the only asset worth more than the redesign.” A reader can agree or disagree with that. Either way they now know you think.

What happened. The honest version. If you have a number, use it. If you do not, say what you do know: “the centre reported enquiries up in the first term, though they also changed their fee structure in the same month, so the signage is not the only variable.” Naming a confound makes every other number on your site more believable, because it shows you are not simply repeating whatever the client said on the phone.

The numbers problem

Most independent creatives never get outcome data. The client does not measure, or will not share, or the project was too small to move anything measurable. This is normal and it is not a reason to invent figures.

Three honest substitutes, in descending order of strength:

1. A client sentence you can quote. Ask for it at handover, while they are pleased, not six months later. The ask that works: “Would you write me two lines about what changed for you? Even just what you were worried about before and whether it happened.” That framing produces a usable quote because it asks about their experience rather than your talent.
2. A continuation fact. “They came back for the packaging four months later.” Repeat business is a real outcome, it is fully within your knowledge, and buyers read it correctly.
3. An operational fact. “Delivered in eleven days against a fourteen-day deadline, printed first time with no plate corrections.” Small, true, and it tells an anxious buyer the thing they most want to know.

What you must not do is write “increased sales by 40%” because the owner said business felt better. Anyone who has run a business knows that sentence is usually unearned, and an unearned number makes the reader discount everything else on the page.

Where people go wrong

The process carousel. Twenty screens of sketches, mood boards, grid overlays and rejected routes. It reads as padding, and worse, it shows the work you decided was not good enough. Two or three process images, chosen because they explain a decision, beat twenty that merely prove you were busy.

Writing about yourself. “I wanted to create something bold and timeless.” Nobody is buying your wants. Replace every sentence beginning “I wanted” with one beginning “the problem was” or “the constraint was”.

The uniform template. If all six case studies have the same six headings, readers stop reading at the second one. Vary the length. A small job can be three sentences.

Doing it with no money

You need no special software. A case study is text and images on a page. Write the text in a plain file first (LibreOffice Writer, Google Docs, the Notes app on a phone) before you touch any layout. If you draft it inside the design, you will write to fill the space.

For the images themselves, Photopea in a browser or GIMP on a desktop will resize and export everything you need; neither costs anything and both open the PSD files an agency might send you back. Export at roughly 1600px on the long edge and save at around 80% JPEG quality: on most portfolio pages that is visually indistinguishable from a 4000px original and loads perhaps four times faster on a phone on a weak connection, which is how a real proportion of your readers will see it.

The limitation worth knowing

A case study only works on somebody who reads. A meaningful share of buyers (art directors scanning forty portfolios in an afternoon, a shop owner on a phone between customers) will look at images and never scroll to the words. The case study is not there to convert them. It converts the slower, more valuable reader: the person deciding whether to spend real money with a stranger. Write it for that person and accept that most visitors will skip it.