The branding agency decision gets made before you meet a single agency

A CMO chooses a B2B branding agency by settling four things inside the company first ... who the company is for, what it argues against, what it will stop saying, and what a win looks like in pipeline ... and then hiring the firm that pushes back on that brief instead of admiring it. Portfolio, process, chemistry and price only sort agencies once that decision exists. Before it, they sort your own uncertainty, and the winner is whichever firm filled the gap most confidently in the room.

That's the whole framework, and it's the opposite of how most selections run. Most run as a bake-off. Five firms, five decks, five different theories of who you are, and a marketing leader trying to pick between them on instinct while a CEO waits for a recommendation with a number attached. The decision you're making feels like a vendor decision. It isn't one. You're choosing whose opinion of your company becomes your company's opinion of itself.

What is the Blank Brief, and why does it decide the agency before the pitch?

The creative brief is agency vocabulary older than any of us, and the good shops take it seriously. Credit where it belongs: the discipline of writing down the audience, the problem and the measure of success came out of agency practice, and firms that insist on one are usually the firms worth talking to.

The Blank Brief is what happens when that document describes your situation and then asks the agency to supply the position. It reads like a real brief. It has the revenue band, the competitive set, the product roadmap, the target segments, the budget. What it doesn't have is a sentence that says what the company stands for and who it's willing to lose.

Every agency you meet will happily fill that in. They have to. They can't pitch without it. And they're good at it, which is the problem, because five capable firms will hand you five coherent answers and none of them came from inside your business. You'll pick the one that sounded best on a Thursday afternoon. Six months later you'll be defending a position you didn't author to a sales team that never heard the argument for it.

This is the same failure as buying a rebrand when the position is what's broken, arriving one step earlier in the process.

Why does this cost a CMO more in 2026 than it did three years ago?

Two things changed, and they compound.

The first is reproduction. Whatever sentence comes out of that engagement doesn't sit in a brand book anymore. It goes into the tools. Your team feeds it to AI writing assistants, your reps paste it into sequences, your site gets rebuilt around it, and within a quarter it's on four hundred surfaces. A position you'd have quietly outgrown in 2022 now gets amplified faster than you can review it. Volume used to cost money, which slowed bad ideas down. It doesn't anymore.

The second is that machines read it. Answer engines ingest what you publish and repeat it back to buyers who never visit your site. If the frame is generic, you get described generically, and you don't get to see it happen. We've written about why an accurate AI description still never turns into a recommendation, and the root cause is upstream of any agency: the company never said anything a machine could quote.

There's a third thing, and it's personal rather than structural. The average CMO tenure is short, and a branding engagement is one of the few line items a marketing leader owns end to end and gets judged on by name. If the choice goes badly, nobody says the brief was blank. They say you picked the wrong agency.

What should you settle before you meet a single agency?

Five decisions. None of them require an agency, and all of them change what an agency can do for you. Give yourself a week and a whiteboard with the CEO and the head of sales in the room.

1. 1Who is this for, narrowly enough to exclude someone. Not the total addressable market. The customer profile where you win more than you lose and the deal cycle is shortest. If your answer includes the word 'and' more than twice, you haven't decided yet.
2. 2What do we argue against. Every position worth having has an opponent: an old way of working, a category assumption, a competitor's premise. If you can't name what you're against, an agency will pick a villain for you, and it will usually be a competitor, which is the weakest option available.
3. 3What will we stop saying. This is the one that gets skipped and the one that reveals whether the decision is real. Name three claims currently on your homepage that you're prepared to delete. A position you're not willing to subtract for isn't a position.
4. 4Who has to be able to say this out loud. If the answer is 'marketing,' the engagement will produce a beautiful document. If the answer includes the SDR making a cold call on a Tuesday, the scope changes, and so does the kind of firm you need.
5. 5What number moves if this works, and by when. Not awareness. Pick something your CRO already reports: win rate against a named competitor, percentage of deals that stall after the demo, average days from first call to second call. Write down the current value.

Answer those five and something useful happens to the agency conversation. You stop asking firms what they'd do and start asking whether they can do this. That's a different meeting, and weak firms hate it.

How do you evaluate agencies once the brief is real?

A framework tells you what to consider. A filter tells you who just got cut. You want filters, and you want them written down before the first call so you're not negotiating with yourself at 6pm on a Friday.

Cut a firm when any of these show up:

• They accept your brief without argument. If you hand five agencies a real position and none of them tells you where it's weak, you've hired an execution shop and called it strategy.
• The work in their portfolio stops at the identity system. Logos, palettes and a tone-of-voice page are real craft, and they don't reach the sentence a rep says on a discovery call. Scope Cliff covers where that boundary usually sits.
• They can't tell you what they'd refuse to change. A firm that would rewrite everything is telling you it hasn't found what's already working. Judge what they'd protect.
• Their process puts the customer interviews after the creative direction. The order is the tell.
• They quote a fixed scope before the position is settled. That price is being set against your uncertainty, and it will move.
• Nobody on the pitch team will be on the engagement. Standard, and still disqualifying at this price.

If you're not yet sure an outside firm is the right shape at all, the four-way comparison of brand strategy firms, messaging consultancies, positioning consultants and fractional CMOs maps what each type actually delivers, and the questions to ask before you sign covers the contract-stage version of this list.

What shows up across 200+ B2B companies?

Here's a number off our own tracking, and it's uncomfortable for an article shaped like this one.

We watch which pages the answer engines pull when a buyer asks how to choose a branding agency. In a 30-day window ending August 5, two pages built exactly like this one ... one written for the CMO, one selling a decision framework ... were each retrieved 15 times and cited zero times. Not once, across the whole window. In the same pull, a page titled 'Rebranding Agency Selection: 7 Filters That Matter' was retrieved 24 times and cited 42.

Why would a machine behave that way? A filter survives being lifted out of context. 'Cut any firm whose portfolio stops at the logo' means something on its own. A framework only works if you read all of it, so there's nothing quotable to carry away.

Your shortlist behaves the same way, which is the actual point. A team can spend six weeks inside an elegant evaluation matrix and still not be able to name which firm got cut on Tuesday and why. Frameworks feel like progress and produce meetings. Filters produce decisions. If you take one thing from this page, take the disqualifier list above and not the five questions.

What does this look like in a real engagement?

This shape repeats often enough that I'll describe it as one company, though it's a composite of several.

A $38M vertical SaaS business, PE-backed, second CMO in three years. Board wants a rebrand because the company looks smaller than it is. The CMO runs a clean process: eight firms, four finalists, scorecard with weighted criteria, references checked. Genuinely good procurement.

Each of the four finalists arrives with a different theory. One says the company is a platform. One says it's a services firm with software attached. One says it's a compliance play. One says it's the modern alternative to the incumbent everyone hates. All four are defensible readings of the same company, because the brief never said.

They pick the platform story. It's the most ambitious, it tests best with the board, and the deck is gorgeous. Eleven months and a large number later, the site is beautiful and sales is still selling the services story, because that's what actually closes. The reps never adopted the new language. Nobody wrote it down as a failure. It just quietly didn't take.

The agency did competent work. They answered the question they were handed. The question was the problem, and it was the company's to answer. If the board's real complaint is that the company looks smaller than it is, that's worth reading against what a rebrand should change and what to leave alone before anyone writes a brief.

What should you do before your next agency call?

Twenty minutes, and you can do it alone before you involve anyone.

Open your homepage. Write down, without looking, the one sentence you'd want a stranger to repeat about the company. Then read what's actually there. If those two things don't match, the gap is the brief, and no agency can close it for you from the outside. If they do match and pipeline still isn't moving, the problem sits somewhere other than the brand, and a rebrand will be an expensive way to find that out.

Then do the cheap version of the whole exercise: run your homepage through the free Brand Signal Score. It grades the same things an agency will spend six weeks discovering, and it takes a few minutes. Walk into the first pitch with that in hand and the conversation changes shape immediately, because you'll be the only person in the room who already knows what's broken.

One more thing worth saying plainly, since this is the part CMOs carry alone. Settling those five decisions before the search protects you as much as it protects the work. When the CEO asks in month nine why the new story isn't showing up in deals, the answer you want is a decision the company made together, with a number attached, that you can point at. A blank brief leaves you holding an agency's opinion and your own name on the invoice.