You've got a positioning document. Someone now has to turn that into type, colour, and imagery.
That translation is real thinking, and almost nobody treats it as its own step. So the decision gets made anyway, implicitly, by whoever's holding the design tool, based on taste. Then three rounds of revisions get spent arguing about something nobody ever wrote down.
Every "I don't know, it just doesn't feel right" you've said in a design review traces back to here. It isn't your fault for not being able to explain it. Nobody gave you the language, because the step where the language gets written was skipped.
When everybody zigs, zag
Marty Neumeier's line, and it's the most useful instruction in visual strategy because it turns an aesthetic question into a research question.
You already did the research. The Category Wall shows you exactly what your category looks like. The move is to study that pattern deliberately and break one significant element of it.
The important caveat: this isn't "go minimal." If everyone in your niche is already minimal, minimal is the pattern, and breaking it means going the other way. The instruction is spot the pattern, then don't be it.
Pair this with one test: write the opposite of each of your brand attributes. If no serious company would ever choose the opposite, the attribute is empty. Nobody is aiming for untrustworthy.
Three words: modern, clean, trustworthy.
Opposites: dated, cluttered, untrustworthy. Nobody picks those. So the words describe the minimum bar, not your brand. We've written all three in first drafts more times than we'd like to admit.
Three words: unhurried, blunt, engineered.
Opposites: urgent, diplomatic, handmade. All three are positions a real company might genuinely choose. That's the sign these words are doing work.
Words before pictures
The temptation is to open Pinterest and start collecting. Don't yet.
Images are easy to agree on and impossible to argue with precisely. You end up with a board everyone likes and nobody can act on, and you've made a decision without noticing you made one. If you can't describe the difference between two directions in words, there isn't a difference. There's two piles of images.
Pull three to five attributes from your positioning. Not aspirational ones. Ones true today. If you have to squint to believe it, it doesn't go on the list.
Pressure-test each word. Could a competitor claim it? If everyone in your category could use it, it's describing your category, not you.
Then build the reference board, and go well outside SaaS. Architecture, print, packaging, film titles, signage, book covers, museum wayfinding. Staying inside your own category is exactly how everything ends up looking like everything else. You already know what your competitors look like. That's the problem, not the input.
Sort into taking and avoiding. The avoid pile does more work than anyone expects. "Not this" is easier to agree on than "this," and just as useful to a designer.
Name three nevers. Three things this brand will never look like. A direction that rules nothing out will collapse the first time someone pushes back.
Teardown
Notion
Around 2018, productivity tools shared a look. Asana, Trello and the rest leaned into bright colour, busy dashboards, and illustrated everything. That was the pattern.
Notion went black and white. Minimal interface, generous space, odd little line illustrations, and a bet on customisation rather than features out of the box.
What's worth understanding is that the minimalism wasn't a taste preference, it was a positioning argument made visually. Every competitor said "we've thought of everything for you." Notion's look said "this is a blank surface, you decide." The visual direction and the product strategy were the same idea.
The postscript is the useful part. It worked so well that the aesthetic became the new pattern, and a wave of tools now look like Notion. Which means if you're launching in that space today, the zag is somewhere else. Patterns move. The method outlasts any particular answer.
What you leave with
What you now have. A written creative direction: three to five true attributes, three nevers, and a reference board split into taking and avoiding.
The test. Hand it to a designer who's never met you. Can they start? If they come back asking what you meant by any of it, it isn't finished.
Where this usually goes wrong. Confusing your taste with the brand's direction. What you personally like and what your buyer needs to feel are two separate inputs. Both are real. Only one belongs in the brief.
What it takes. Three to five days. If you're doing this alone, the hard part isn't collecting references, it's the adjectives. Hand your list to one person who doesn't work there and ask them to cross out anything generic. Don't defend the words. Watch what survives.
