Do we need a branding agency or a marketing agency?

A branding agency shapes what your company says and how it looks. A marketing agency takes what you already say and puts it in front of buyers. If your leadership team can write one page today on why a buyer should change, why now, and why you, then buy the marketing agency and buy as much of it as you can afford. If they can't write that page, neither firm is your first purchase.

That's the answer. Most founders still get it wrong, because the proposals don't sort themselves. Both firms open with a discovery phase. Both include a line item called messaging. Both show you logos you recognize. The sorting has to happen on your side of the table, before anyone pitches.

What does each one actually sell?

A branding agency sells articulation and appearance. Identity system, naming, tone of voice, a brand book, often a website. What you get is a set of artifacts that describe who you are, and they sit still until someone uses them.

A marketing agency sells motion. Demand generation, content, paid media, SEO and AEO, lifecycle email. What you get is activity that runs, and it stops running the month the retainer stops. If you want the long version of that job description, we wrote it in what a tech marketing agency actually does.

Here's the part neither proposal says out loud: both firms are execution shops. One executes on how you appear, one executes on how far you reach. Neither is a strategy department, and a portfolio can't tell you where their scope ends. That boundary is the whole game when you're choosing between branding firms, and we mapped it in how to choose a B2B branding agency that moves pipeline.

Why does messaging show up in both proposals?

Because neither firm can start without one. The branding agency needs a message to dress. The marketing agency needs a message to distribute. When you don't hand one over, both write their own, and they write it as a phase inside a bigger deliverable rather than as the deliverable itself.

Watch what that produces. The branding agency delivers a messaging framework inside the brand book in March. In October the marketing agency writes a campaign brief with its own message hierarchy, because the brand book didn't answer the questions a campaign asks. Now you own two messaging documents, with the same name, written by two vendors, and neither one outranks the other. Your sales team read neither, so discovery calls run on a third message that lives in a rep's head.

There's a tell you can read in the proposal itself, before you meet anyone. Find where messaging sits in the timeline and how much of the fee it carries. If it's week two of fourteen, bundled into discovery and priced at nothing, it's a phase on the way to the real deliverable, and phases get compressed the moment production runs late. If it's the spine of the engagement with its own sign-off gate, they intend to settle something.

Nobody named the referee. That second messaging document is where the money actually leaks, and it leaks quietly, because every individual deliverable looks fine. The failure only shows up in the gap between them.

Which gap do we actually have?

Three checks, and you can run all of them this week without buying anything.

1. 1The one-page test. Ask four people on your leadership team to write, separately and without conferring, why a buyer should change, why now, and why you. Then read the four pages side by side. Four different answers means the decision was never made, and no vendor can be bought out of an unmade decision. They can only render it.
2. 2The pattern test. Pull two quarters of performance by channel. If results vary channel to channel, distribution is the variable and a marketing agency has real room to work. If everything lands in the same flat band no matter the channel, the spend or the creative, then the message is the constant. We walk that distinction through in how to tell if your messaging is broken rather than underperforming.
3. 3The stranger test. Cover your logo, show your homepage to someone outside your category, and ask who it's for. If they can't answer in a sentence, an agency is going to inherit exactly that blank and fill it with the category average. Our Brand Signal Score runs a version of this on your homepage and tells you which half is failing, the narrative half or the conversion half.

Notice what the three checks have in common. None of them ask which firm is better. They ask which layer is broken, because that's the only question a vendor choice can answer.

What does buying the wrong one first cost?

Buy distribution on an undecided message and you don't get nothing. You get something worse. You get volume of the wrong thing, plus a year of clean-looking data that says the channels failed. Then you fire the agency and hire the next one, which is the loop we described in why every marketing agency you hire fails.

Buy branding on a message that's already decided and you get a handsome restatement of what you knew. That one's survivable. It's slow and it's expensive, and nothing downstream breaks.

We ran the expensive version of this on ourselves, in the opposite direction. PitchKitchen has published more than three hundred articles, and on the tracked branding vocabulary we register under one percent across 651 recorded AI answers. The argument isn't weak. We'd simply written narrative identity for years while our buyers were typing branding agency. We owned the decision and lost the search, because we'd picked a label our market doesn't use.

Here's the part worth planning for: at this size most companies end up buying both firms eventually. The order is what you actually control, and the order decides whether the second purchase inherits a settled argument or inherits a vacuum. A marketing agency handed a real position multiplies it. A marketing agency handed a brand book that stops at tone will write its own position by Thursday, and now you're back to two documents.

What should we settle before we sign either one?

Four things, and all four belong in the contract rather than the kickoff deck.

1. 1Name the source of truth. One document holds the position, and it has one owner inside your company. Write down which document that is by name before the first invoice.
2. 2Make everything else derive from it. Every downstream artifact, brand book, campaign brief, sales deck, website, is derived from that document rather than inspired by it. Those two words behave very differently when a deadline gets tight.
3. 3Set the seam. If you're buying both firms this year, name the date the branding work becomes the marketing agency's input, and name who signs off that it's ready to hand over. Unowned seams are where the second messaging document gets born.
4. 4Decide what you keep. Ask each firm what you own on day one hundred if the relationship ends, and get the answer in writing. A decision you own outlives the vendor. A campaign you rent doesn't.

Where this leaves a growth-stage B2B

Between five and seventy-five million in revenue, you're usually funding one of these firms this year, not both. The good news is that the sequencing question answers itself once you've run the three checks. Decide first, distribute second, and refuse to let a vendor make the decision by default inside a deliverable you'll never read closely.

If the one-page test comes back with four different answers, you don't have an agency problem yet. You have an unsettled position, and that's the thing to go fix. We named that layer narrative identity, and we keep it separate from the visual layer for exactly this reason, which we lay out in brand identity versus narrative identity.

PitchKitchen builds Magnetic Messaging Frameworks for founder-led B2B companies in the $5M-$75M range. Greg Rosner, author of StoryCraft for Disruptors, built the framework so that one decided argument can be handed to any vendor, any rep, and any AI tool your team writes with, and survive all three.