Constellations, patent pending

What 500 People See · Process

What a Brand Audit Misses

An inventory grades the work against a document nobody outside the company has read. Useful, and one step short of the thing you needed to know.

Quick answer

A standard brand audit inventories what exists and grades how far it has drifted from the style guide. Website templates, decks, print, social, packaging, all of it collected and marked on or off brand. That is useful and it is incomplete, because it only measures the work against a document. An audit that measures something real adds one step: sample the material so it represents the full spectrum of what the brand puts into the world, then test that sample with the people closest to the brand and, where it matters, with the audience. Consistency with the guide is a filing question. Whether the brand is being read the way you intended is the audit question.

What a normal audit gives you

Run a conventional audit and you end up with a list, which is worth having.

You find everything that has been made over the years. Website templates. PowerPoint decks that shipped from three different departments. Print material. Trade show material somebody's cousin designed in 2019. You lay it out and evaluate how on or off brand each piece is, and the drift becomes visible for the first time.

That surfaces real decisions:

  • Which pieces are on brand but simply outdated and need refreshing?
  • Which pieces have drifted far enough that they should be pulled back and redesigned?
  • Where has the drift produced something better than the guideline, so the guideline should be updated to accommodate it?
  • Do the guidelines need adjustment anyway, because of how the industry or the company has moved since they were written?

How deep you go depends on the size, scope and purpose of the audit. But the shape is the same. Gather everything, evaluate it against the standard, decide what to fix and what to formalize.

The question the inventory can’t answer

Everything above measures the work against a document that a small group wrote at a particular moment.

It doesn't tell you whether the document was right. It doesn't tell you whether the pieces that score as perfectly on brand are the pieces that read as the company to anyone outside the marketing team. A brand can be internally consistent, well executed, fully compliant with its own guidelines, and still be communicating something nobody intended.

Figure 1 · Two measuring sticks, one piece of work

The same asset graded against the style guide and against audience response One asset splits into two evaluations. On the left it is checked against color, typeface, logo lockup and clear space and marked fully compliant. On the right the same asset produces a scattered field of responses with no cluster, marked as reading as anyone. One piece of the brand MEASURED AGAINST THE GUIDE Color palette Typeface Logo lockup Clear space FULLY COMPLIANT MEASURED AGAINST THE PEOPLE IT’S FOR READS AS ANYONE No cluster. Nothing agreed on.
A compliance score and a perception result are answers to different questions, and only one of them is about the people the brand is for. The left panel can be perfect while the right panel is empty.

Consistency is a floor. A brand can be consistently wrong, and the more disciplined the organization, the more efficiently it distributes the wrong thing.

How to run the measured version

The measured audit uses the inventory as its input and adds a response layer on top.

  1. Collect everything. Same first step as the standard version. The full list of what exists.
  2. Sample it deliberately. Pull the pieces you consider representative of the brand and make the spectrum as broad as you can. Include the strong work and the drift. Include the material that makes you wince, because it is out there being seen either way and you need to know how it reads.
  3. Test the sample with the people closest to the brand. Send it to the employees. Ask which pieces feel most representative of the brand and its values, and which feel off. This is where the surprises live, because the people producing and using this material every day have never been asked to respond to it as a set. That is what a perception test is for.
  4. Put the results side by side. The pieces that scored high with strong agreement, and the pieces that clearly read as off. Compare the two collections directly. This is the step teams skip, and it is the one that does the work.
  5. Find what each collection has in common. Comparative analysis across the high scorers and the low scorers, which is a genuinely good use of AI, since finding patterns across a body of qualitative material is what it is good at.

Figure 2 · Contrast is what makes the line visible

High-scoring and low-scoring collections held against each other Two groups of eight asset tiles. The left group is marked as scored high with strong agreement and resolves to a summary reading what the brand actually is. The right group is marked as clearly read as off and resolves to a summary reading what to stop making. A dashed line runs between them. The step teams skip: hold the two collections up against each other. SCORED HIGH, STRONG AGREEMENT What the brand actually is CLEARLY READ AS OFF What to stop making THE LINE
What people rejected, held next to what people responded to, makes the dividing line visible in a way a pile of approved material never does. The positives alone give you a mood board. The pair gives you a rule.

Those two answers give you something a style guide can't: a description of what the brand is, in terms of the elements people respond to, and a description of what to avoid.

Where the brand is customer-facing in a way that matters, run the same sample past the audience as well. Internal alignment tells you whether the company knows itself. Audience response tells you whether the market agrees, and where the two part company is worth knowing before either one drives a decision.

What comes out the other side

The output isn't a compliance report. It is a set of visual evidence a team can point at.

One company that ran this on themselves, Keen Learning, took the material they had in the world, built a few boards out of it, and gave the whole company a gut check. What came back was strong alignment on what looked appropriate for the brand and what looked off. They could see it, point to it, and agree on it in one pass.

The decision they made from there was to get back to the brand guidelines they had drifted from, and to keep using only the material that worked. A Rutgers arts center pointed the same move at a different question during a year of new leadership, testing where a sceptical staff actually agreed before setting direction. That is an ordinary outcome and a good one. The difference is that the whole team arrived at it together, looking at the same evidence, instead of receiving it as a directive from marketing.

Everyone agreed, and they knew why, because they saw why.

When an audit should turn into something bigger

Sometimes the test comes back and the answer isn't drift.

If the material that scores highest with employees is nothing like the material that scores highest with customers, the brand has a positioning problem, not a consistency problem. If nothing clusters anywhere, if the responses scatter across everything, that is a signal that the brand isn't carrying a clear idea to anyone, which no amount of guideline enforcement will fix.

Figure 3 · Three things the result can be

Three possible response patterns from a measured audit and what each means Three panels. The first shows one tight cluster of responses, read as drift. The second shows two clusters far apart, read as a positioning problem. The third shows responses scattered evenly with no cluster, read as a brand not carrying an idea. Drift One tight cluster. Fix the material. Update the guide. A positioning problem Two clusters, far apart. Not a consistency problem. No signal Nothing clusters anywhere. The brand isn’t carrying an idea.
Only the first panel is the problem most audits are commissioned to solve. The other two are worth far more than a color-coded spreadsheet, and neither is visible until somebody responds to the material.

Those findings hurt, and they are worth far more than a color-coded spreadsheet of compliant and non-compliant assets. Better to learn it from an audit than from a rebrand that goes live and lands the same way the old one did.

Frequently asked questions

What should a brand audit include?

At minimum, a complete inventory of brand material and an evaluation against current guidelines. To be worth the effort, add a measured layer: a representative sample tested with employees and, where relevant, the audience, comparing what reads as the brand against what reads as off.

How do I do a brand audit myself?

Collect everything the brand has produced, sample it broadly, test the sample with your team and your audience, then compare the high scorers against the low scorers and identify what each group has in common. Update the guidelines to reflect what is true, not just what was written.

Who should be involved in a brand audit?

More people than usually are. The employees living with the brand every day have a read on it that the marketing team cannot have, and including them produces both better data and an easier rollout afterward.

How often should you audit a brand?

Whenever drift is visible, before a rebrand is considered, and before any major campaign builds on assumptions about how the brand is being read. An audit that only happens when a rebrand has already been decided is documentation, not diagnosis.

Pillar: What 500 People See (Process). Target keyword: brand audit. Search intent: “brand audit,” “how to do a brand audit,” “brand audit template.”