10 August 2026

Read the brand, do not invent it

The best branding work I have done was noticing, not designing. Most people already have an identity — it is just inconsistent.

  • branding
  • design
  • process

There is a version of branding work where you go away for two weeks, come back with three directions, and the client picks the one that looks most like what they already had.

There is a faster version that produces better work, and it starts with the assumption that the brand already exists and nobody has written it down.

It is usually already there

Anybody who has been posting for a year has an identity. It is scattered across a channel banner, a profile picture, the colour of a mascot's shirt, the way they write captions, the three words they say constantly.

None of that is documented. All of it is real, and their audience already recognises it.

So the first job is not designing. It is reading. Go through everything they have made and write down what is actually consistent:

  • What colours keep appearing, including in photos and artwork, not just in logos
  • What the typography on their own thumbnails is trying to be
  • The register of their writing — do they use full sentences or fragments, are they funny, do they swear
  • What they call their audience
  • What they never do

That list is a brand guide. You did not invent any of it and it is more accurate than anything you would have invented.

Why this produces better results

Two reasons, and the second one matters more.

The first is speed. You skip the exploration phase because the exploration already happened, in public, over a year, with real feedback.

The second is recognition. When you show someone a system built out of their own existing signals, they recognise themselves in it immediately. There is no selling, no "let me walk you through the thinking". They just say yes, because it is theirs.

Compare that to presenting three invented directions. Now you are asking someone to choose an identity, which is a genuinely hard thing to do, and the failure mode is that they pick the one that is safest, and you have made them more generic than they were before you arrived.

Where invention is still needed

Reading gets you eighty per cent. The rest is real work:

  • Making it consistent. Everyone's existing identity is inconsistent, and consistency is most of what "professional" means visually.
  • Filling the technical gaps. A photograph cannot be a favicon. A screenshot cannot be a print file. Someone has to make the mark that works at sixteen pixels.
  • Deciding what to drop. There will be things in the existing mix that are holdovers from an older version of them, and part of the job is saying so.

The test

Show it to them without explanation. If they need the rationale, you invented too much.

The best reaction to branding work is not "that's clever". It is "yeah, that's me" — said slightly impatiently, like you have wasted their time telling them something obvious.

That is the sound of a job done right.

← All writing