What brand drift actually costs
A documented operational tax.
Scene illustration — in build
Brand drift looks like nothing. That’s why it’s expensive.
A new hire writes the next customer email. They don’t know the voice spec exists. A team member updates the sales deck and picks a colour that feels close-enough to the brand palette. The website hasn’t been touched in eight months because no one’s noticed it ought to be. Each piece, in isolation, is a small thing. None of them is a crisis. There’s no incident report. Nobody is fired. The brand simply continues, slightly off in three different directions, and the cost is paid by customers in the form of trust they didn’t quite extend.
The cost is documented. The research has been there for years. The strange thing is how rarely it shapes the way founder-led teams operate their brand.
The documented cost
Eighty-one percent of companies produce off-brand content regularly. That figure comes from Lucidpress’s 2019 State of Brand Consistency research, which surveyed more than 400 marketing leaders.1 The phrase “off-brand content” is doing work here: it doesn’t mean catastrophically wrong, it means systematically inconsistent. A subtitle that contradicts the voice spec. A logo variant deployed in a context the brand guidelines explicitly warned against. A customer email that drifted into a register the founder would never write in herself.
The 81% figure is striking because it isn’t a story about bad teams. It’s a story about teams operating without a system. Off-brand content is a default state, not a failure mode — which is exactly why discipline-based interventions don’t fix it.
The same Lucidpress research surfaced the operational consequence: marketing leaders in those off-brand-producing companies spend roughly 20% of their working week correcting off-brand materials.1 Twenty percent. One full day out of every five. The day doesn’t appear on anyone’s calendar as “rework day” — it’s distributed across mornings and Slack threads and last-minute edits before a campaign ships. But the time is real, and the time is the founder’s, or her senior person’s, or a creative director’s whose attention was meant to be on the next strategic move and is instead on cleaning up the last tactical one.
The Renderforest 2024 Brand Guidelines Enforcement Study widened the lens. Ninety-five percent of organisations have written brand guidelines.2 Fewer than 30% actively enforce them. That’s a sixty-five point gap between documentation and operation. It means most brand-drift problems are not because the guidelines were missing. The guidelines are usually thoughtful — somebody did the work to make them exist. They’re in a Notion folder, or a PDF on Drive, or a Figma file with proper layers.
The guidelines exist. The work of running them does not.
The interesting research isn’t the 81% or the 20% in isolation. It’s what happens when small drifts compound across the customer’s experience. Salesforce’s 2024 State of the Connected Customer survey found that 76.2% of customers abandon a brand after three or more inconsistent interactions across channels.3 The number is brutal because three is not a lot. A customer might interact with a brand five or six times before deciding. Three inconsistencies inside that window — a website that says one thing, a sales conversation that frames it differently, a customer-support reply written in a third register — is enough to push the majority of those customers toward the exit.
The drifts don’t have to be dramatic. They have to be perceptible. And the cumulative effect, across a thousand customers and a year of operation, is the kind of churn pattern that shows up in the dashboard with no obvious cause. The cause was the drift. The drift was paid for in trust.
Why discipline doesn’t solve this
The most common response to brand drift is a discipline-based intervention. The leadership team agrees that brand consistency matters. A new policy is announced. A weekly brand review meeting goes on the calendar. The team is asked to be more careful.
For three weeks, things tighten. By week six, the review meeting starts running over. By week ten, it’s been moved to alternate weeks. By month four, it’s been dropped, and the team is back to making the same micro-decisions in the same micro-drifting ways.
Discipline-based interventions fail because they ask people to remember to do something the system isn’t doing for them. Asking a content writer to remember the voice spec while drafting a launch announcement is asking her to context-switch out of the writing she was hired to do. Most people, most of the time, won’t — not because they’re undisciplined but because the discipline costs cognitive bandwidth that the work itself needs. What works isn’t more discipline. It’s less reliance on discipline.
Reading this far? We send essays like this once a month — no drip funnels, no autoresponders, no upsell. Just the kind of analysis you’re reading now.
Get monthly Notes from RTSNThe system, not the discipline
What replaces discipline is operation. Not better adherence to guidelines — actual systems that run the guidelines so the team doesn’t have to remember them.
That looks like a voice spec wired into the team’s writing tool, so off-brand drafts get flagged before they ship. It looks like a palette enforced at the design-tool level, so off-palette use surfaces immediately. It looks like a content cadence that runs on its own rhythm with AI-handled drafting, human-handled judgment, and documented escalation thresholds. It looks like an audit trail you can query in seconds — what was decided, when, why.
The McKinsey Design Value research found that design-mature companies returned 32% more revenue growth than peers over a five-year window.4 That’s an industry-benchmark figure, not a promise. What it tells us is that the difference between a brand that compounds and a brand that erodes is operational, not aesthetic. The companies that retain trust over years aren’t the ones with the most disciplined teams. They’re the ones with the most documented systems.
The guidelines exist. The work of running them does not. That’s what we build.