Week 1
Map what happens now
How work actually gets approved today, which is usually undocumented and rarely what anyone believes it to be.
Creative direction · Creative governance · Sydney
Creative governance
Every brand that has come apart did so one reasonable exception at a time, approved by somebody with a deadline and no authority to say no. Governance is just deciding who can say no, and when.
Grab the snowball. Give it a spin, and watch it grow as you scroll. Just like the standard.
1named owner who can say no
3levels of review, not one queue
48 hoursa turnaround people can plan around
Fewer meetingsnot more process
The short answer
What is creative governance?
Creative governance is the set of decisions about who approves creative work, at what level of risk, and how quickly. It is not more process. Done properly it is less: most work stops needing approval at all, and the review that remains is fast and by somebody qualified to give it.
It exists because consistency is not a design problem past a certain size. It is an authority problem. If nobody can say no, and nobody can say yes quickly, people either wait or proceed, and both outcomes damage the brand.
It is the part clients most often skip and most often need, so this page exists to explain what would be skipped. Because our Snowball SEO platform automates the search heavy lifting other agencies bill by the hour, more of your budget goes into the work rather than into the administration of it. It sits inside our wider creative direction service.
The difference
How much review does work actually need?
Almost none, for most of it. The mistake in nearly every governance model is treating a social post and a television commercial as the same decision.
What good governance looks like
- Most work needs no approval, because the template already decided
- One named reviewer for the middle tier, with a stated turnaround
- A real decision forum only for genuinely new or risky work
- A documented answer for the awkward cases, so they are not escalated
What bad governance looks like
- Everything through one queue and one person
- Approval by committee, which converges on the least objectionable option
- No stated turnaround, so people stop asking and start guessing
- Rules with nobody empowered to grant an exception
Governance encodes decisions that already exist. If they do not, that is visual direction first, and governance afterwards.
The cost
What drives the price of governance work?
Organisation size and how many parties are involved. Governance for one team is a conversation; governance across regions, suppliers and franchisees is a design problem.
A review model
The decisions
Who reviews what, at what risk level, and how fast, written down and agreed with the people it affects.
Model and tooling
Made workable
The same, plus where requests go, how they are tracked and what the turnaround commitment is, so it survives a busy week.
Governance at scale
Multi-party
For businesses with regions, franchisees or many suppliers, where the model has to work without everyone knowing each other.
We quote after a free consult and a look at how work currently gets approved, which is often the first time anyone has written that down.
What moves the number, in order
- How many people and teams make creative work
- Whether external suppliers, franchisees or regions are involved
- How much is regulated or carries legal risk
- Whether tooling and tracking are in scope
- How much existing process has to be unpicked
- Whether training the reviewers is included
Governance should remove meetings
If the outcome adds a meeting, it has failed. The point is to move most decisions out of meetings entirely, by making them in advance and encoding them in templates and rules.
The test is simple: after the model is in place, how many things need a conversation that previously did? If the answer is more, the model is describing an approval chain rather than a governance system.
So we design for the smallest amount of review that keeps the brand intact, and we measure success by how little of it is needed.
The scope
What you receive
A model people can follow, and the authority to make it real.
The review model
What needs review, at what level, and what does not, written so somebody with a deadline can work out where their thing sits.
Named roles
Who decides at each level, with named people rather than job titles, because a title cannot approve anything.
Turnaround commitments
How fast each level responds, stated. Without this people stop asking, which is worse than a slow queue.
The exceptions process
How to get a considered yes to something outside the rules, since a system with no exception path gets ignored entirely.
A supplier version
The same model expressed for external partners, who otherwise operate outside it by default.
A review checklist
What a reviewer actually looks at, so review is consistent rather than dependent on who happens to be looking.
The return
What happens when nobody owns it?
The brand does not collapse. It erodes, gradually and invisibly, and the cost shows up as a rebrand two years later that would have been unnecessary.
The cost is invisible until it is a rebrand
Nothing dramatic happens when a brand has no governance. Assets drift slowly, each one defensible, until somebody new arrives and observes that the brand looks tired and inconsistent.
The response is usually a rebrand, which is expensive, disruptive and often unnecessary: the identity was fine, it was simply never governed, and the new one will drift the same way within three years for the same reason.
That is the argument for this work, and it is a hard one to make in advance, which is why the page exists at all.
The process
How governance gets set up
Three to five weeks, mostly spent agreeing who has authority rather than designing process.
Week 1
Find the bottleneck
Almost always one person, informally, who never agreed to the role and cannot sustain it.
Week 2
Tier the work
What needs no review, one reviewer, or a real decision, decided by risk rather than by seniority of the requester.
Week 3
Name the people
Actual names at each level, with turnaround commitments they have agreed to rather than been assigned.
Week 4
Write the exception path
How to get a considered yes outside the rules, because a system without one is routed around within a month.
Week 5
Roll out and review
Agreed with the people it affects, then revisited after a month, because the first version is always slightly wrong.
The brief
What to ask before you set up governance
Governance is bought to increase control and works by reducing it in most places. These questions test for that.
How much work will need no review at all?
It should be most of it. If the model reviews everything, it is an approval chain, and it will be routed around by the second busy week.
Who is named, and have they agreed?
Names rather than titles, and people who have consented to the role and the turnaround. An assigned reviewer who never agreed is a bottleneck with a job title.
What is the turnaround commitment?
Stated, and short. Without one, people stop asking and start guessing, which produces exactly the drift the model was meant to prevent.
What is the exception path?
There must be one. A system with no way to get a considered yes to something unusual gets ignored entirely, and then you have neither rules nor exceptions.
We are happy to answer all four and to map how approvals currently work before you commit. Governance encodes decisions; if those are unsettled, start with visual direction.
Start here
Ready to decide who decides?
Tell us how creative work gets approved today, as it actually happens rather than as the process says. The free consult is a working session, not a sales call, and you leave with three things whether or not you book us.
- A map of how approvals actually work now, which is usually undocumented
- The bottleneck, named, and what it is costing in time
- A fixed price for the model, with tooling costed separately
Good questions
Creative governance FAQs
What is creative governance?
It is deciding who approves creative work, at what level of risk, and how quickly. Done properly it is less process rather than more: most work stops needing approval at all, and what remains is fast and reviewed by somebody qualified.
Is this not just adding bureaucracy?
It is the opposite when done well. If the outcome adds meetings, it has failed. The test is how many things need a conversation afterwards that did not before, and the answer should be fewer.
How much work should need review?
Most of it should need none, because it was built from templates that already encode the decisions. A smaller share needs one qualified reviewer, and only genuinely new or risky work needs a real decision.
How much does governance work cost?
We quote after mapping how approvals happen today. What moves it is how many people and teams make creative work, whether suppliers or regions are involved, how much is regulated, and whether tooling and reviewer training are in scope.
Who should be the reviewer?
Somebody who has agreed to the role and can commit to a turnaround, named rather than designated by job title. An assigned reviewer who never consented is a bottleneck, and it will be discovered during a busy week.
What if there is no exception path?
The system gets routed around within about a month. Every set of rules meets a case it did not anticipate, and if there is no way to get a considered yes, people will simply proceed without asking.
Do we need this if we are small?
Probably not yet. Governance becomes worth formalising when more than a handful of people make assets, or when external suppliers are involved, because that is the point at which informal agreement stops working.
What happens without governance?
Nothing dramatic, which is the problem. The brand erodes gradually, each deviation defensible, until somebody concludes it looks tired and proposes a rebrand. That rebrand is frequently unnecessary and will drift the same way within three years.