A brand programme is delivered. The work is good — considered, well-argued, properly made. Guidelines are issued. Eighteen months later the organization looks approximately as it did before, with a new logo applied inconsistently across materials that contradict each other.
This is such a reliable outcome that it is worth treating as a systems problem rather than a series of unfortunate cases.
The identity is usually not the failure point
In our experience the design work is rarely the cause. Most professionally delivered identity systems are coherent and would hold up under sustained use. What fails is everything that has to happen afterwards, in an organization that has just spent its budget and its patience and now regards the project as complete.
Three mechanisms account for most of it.
Nobody owns enforcement. Guidelines describe correct usage; they do not compel it. In the absence of a named person with the standing to reject non-compliant work — including work produced by a senior executive’s preferred agency — the system degrades at the rate that exceptions are granted. The first exception is always reasonable. It is also the precedent.
The system was specified for the launch, not for the work. Identity programmes are designed against a showcase: a campaign, a website, a headquarters. The actual daily demand is a procurement pack, a regional price list, a recruitment advertisement in a language the system was never tested in, and a partner co-branding request. If the system has no answer for these, teams invent one, and the invention becomes the standard because it is the only thing that exists.
Nothing was decommissioned. The old assets remain in circulation — templates on shared drives, printed collateral, third-party listings, partner materials. Two years on, both systems are in use and the organization has less coherence than before the programme began.
Year two is where the cost appears
The reason this surfaces in the second year rather than the first is straightforward. In year one, the launch team is still assembled and attention is high. Compliance looks excellent. In year two, the team has dispersed, the sponsoring executive has moved on or moved subject, and the system meets the long tail of real requests with nobody left to arbitrate them.
By the time it is visibly failing, the organization has generally concluded that the identity was wrong, and begins considering another programme. This is the expensive misdiagnosis. A second identity administered the same way will fail the same way.
What to specify instead
The commissioning brief is where this is fixed, not the design.
- Name the owner before the work starts, with the authority to decline non-compliant output, and place them somewhere the authority is real.
- Specify the system against the twenty least glamorous applications, gathered from the business rather than imagined. If it cannot produce a clean procurement document and a compliant regulatory notice, it is not finished.
- Fund the rollout as a separate line, sized at a meaningful fraction of the design cost. Programmes that fund creation and assume implementation are funding half a project.
- Plan decommissioning explicitly — what is withdrawn, from where, by when, and who confirms it.
- Set a governance cadence: a quarterly review of what has actually been produced, with the authority to correct it.
The test worth applying
Before approving a brand programme, ask who will reject the first piece of non-compliant work produced by someone senior, and what happens when that person objects.
If the answer is unclear, the programme will fail in year two regardless of the quality of the design. If the answer is clear and credible, the identity itself matters considerably less than anyone involved will want to admit.