How to Choose a Branding Partner in 2026

Most branding partnerships fail before anyone opens a design file. They fail in the brief.

Here is the pattern, and once you see it you will see it everywhere.

A company knows something is wrong. Sales is having to explain too much. The website feels dated. The deck is dusty. New hires describe the company three different ways at dinner. So leadership writes a brief that says "we need to refresh our brand," sends it to eight agencies, gets eight beautiful decks back, picks the one with the best chemistry, and spends six months and a lot of money making the surface prettier.

Then the underlying problem is still there. Because the underlying problem was never the logo.

The problem was that the company had become hard to interpret. Which used to be a messaging problem and is now also a machine problem, because the systems summarizing you for your buyer have to interpret you first.

I have watched this happen from both sides for twenty years of watching this. So this is not a checklist for procurement. It is what I would tell you if you called me and I had no chance of winning the work.

First, diagnose before you shop

You cannot choose a partner until you know what is broken. Most companies skip this and it costs them the entire engagement.

There are four different problems that all feel identical from the inside, and they need four different kinds of firm.

Problem 1: You look wrong. The identity is dated, inconsistent, or was built for a company you no longer are. This is an identity problem. Design studios and identity specialists solve it. It is the cheapest of the four and the most commonly over-diagnosed.

Problem 2: You sound wrong. People understand what you do but not why it matters, or why you and not the other guy. This is a positioning problem. It is a strategy engagement, and no amount of visual work fixes it.

Problem 3: You are structurally confusing. Four names for the same product. Sub-brands that fight each other. An acquisition nobody integrated. This is brand architecture. It is specialist work and it is expensive because it is mostly organizational, not creative.

Problem 4: The market has not caught up to what you have become. The business changed, the story did not. This is repositioning, and it is a leadership decision wearing brand clothes.

Write down which one you have. If you cannot pick one, that is diagnostic information too, and it means your first engagement should be small and exploratory rather than a full rebrand.

The test: if you cannot articulate the problem in two sentences without using the word "refresh," you are not ready to hire anyone yet.

Second, decide who owns this internally

Branding projects die from committee more than from bad agencies.

You need one person who can make the final call. Not a consensus. A person. Brand work requires choosing what you are not, and groups are structurally incapable of choosing what they are not, because every person in the room protects a different thing.

Before you talk to a single firm, answer these:

• Who has final approval, by name
• Who has veto power, by name, including anyone on the board who will surface in month five
• Who is providing the day to day input
• Which stakeholders need to be consulted versus merely informed, and who decided that

That last one prevents the most expensive failure mode in this entire process: a senior person appears at the eleventh hour with opinions nobody has heard yet, and the work resets. Every agency has this scar. Ask them about it in the pitch and watch them flinch.

Third, write a brief that is actually useful

A good brief is short and specific. A bad brief is long and vague and reads like it was assembled from four departments, because it was.

Include these six things.

1. The business situation, honestly. Not the version for the press release. What is actually happening. "Growth flattened last year and we think the reason is that we are being compared against the wrong category" is a brief. "We want to be seen as innovative leaders" is not.

2. What has to be true in two years. Not brand goals. Business goals. Revenue, category position, customer type, valuation, whatever it is.

3. What is off the table. The name you will not change. The equity you will not abandon. The founder's attachment you will not fight. Every constraint you hide will surface later and cost more.

4. Your actual audience, narrowed. "Enterprise decision makers" is not an audience. If your audience list has more than three entries, you have not made a decision and the agency will have to make it for you.

5. The budget range. Yes, tell them. Agencies who know the number will scope honestly. Agencies who do not know it will guess, and half of them will guess wrong in a direction that wastes everybody's time. Withholding the budget to get a "clean" proposal is a procurement habit that produces worse proposals.

6. Your decision timeline and process. How many rounds, who is in them, when you decide.

That is it. Two pages. If your brief is thirty slides, you have written a strategy document, and you are now paying an agency to read it back to you.

Fourth, build the shortlist

Three to five firms. Not eight.

Eight means shallow evaluation, exhausted internal teams, and seven agencies doing unpaid work, which selects for firms with capacity to burn rather than firms doing their best work.

Where to look, in order of signal quality:

Their published work in your problem, not your industry. Industry experience is overrated and agencies know you want it, so they lead with it. What matters more is whether they have solved your kind of problem. A firm that repositioned a complex energy business can probably reposition a complex logistics business. A firm that has done twelve companies in your exact category may just give you the thirteenth version of the same brand.

Case studies with numbers and conflict. Look for what the business result was, and what was hard. Case studies that are pure celebration are marketing. Case studies that describe a real disagreement and how it resolved are evidence. Apply this to us too. That link goes to ours.

Referrals from clients who left. This is the highest-signal call you can make and almost nobody makes it. Ask the agency for a client whose engagement ended. If they cannot produce one after twenty years in business, something is off.

Who their people are. Look at the team page and then look at those people on LinkedIn. Are they the ones doing the work, or is there a layer of directors between the pitch and the production?

Our companion piece, the top branding companies for 2026, breaks down which firms are built for which problem, including where we fit and where we do not.

Fifth, run a process that tests the right things

Most agency selection processes test presentation skill. Presentation skill has almost no correlation with the quality of the work you will get.

Here is what to do instead.

Kill the spec work. Do not ask for creative concepts in the pitch. You will get work made by whoever was free that week, based on no research, designed to win a meeting rather than solve your problem. It teaches you nothing except which firm has the most idle designers. It is also the single biggest reason good small firms decline to compete.

Pay for a paid diagnostic instead. Give your top two or three a small, paid engagement. A workshop, a positioning sprint, an audit. Real money, real scope, two to four weeks. You will learn more about how a firm thinks in one paid session than in three pitch meetings, and you will find out whether they can disagree with you productively before you are eight months and six figures in.

This is the single best change you can make to your process. It costs a little and it de-risks a lot.

Meet the working team. Not the pitch team. If the people in the room are not the people doing the work, that is not automatically disqualifying, but you need to know, and you need it in the contract.

Give them a real problem in the room. Not a hypothetical. Something you are genuinely stuck on. Watch how they think out loud, what they ask, and whether they are comfortable saying "I do not know yet."

Sixth, ask the questions that actually separate firms

Every agency has rehearsed answers for the standard questions. These are the ones that produce unrehearsed answers.

"Who specifically will do this work, what percentage of their time, and what else are they on?"The most common gap in this entire industry is between the team that sells and the team that delivers. Get names and percentages. Put them in the contract.

"Tell me about a project that went badly."Everyone has them. A firm that cannot name one is either lying or has not been trusted with anything hard. Listen for whether they take responsibility or blame the client. Then notice: if they blame the client here, they will blame you later.

"What did you talk a client out of?"This tests whether they have a point of view or a service menu. If the answer is nothing, you are hiring a vendor, not a partner. Vendors execute your bad ideas beautifully.

"Show me a case study where the business result changed, and tell me what the number was before."Anyone can show growth after a rebrand. The question is what the baseline was and what else was happening. Firms that know their clients' business will have the answer. Firms that shipped a logo and left will not.

"What happens after you leave?"You are looking for a system your team can run: guidelines, source files, an asset library, naming conventions, and someone internally trained to hold the line. If the answer is a PDF, you will be back here in two years.

"Can an AI system describe your last three clients correctly?"Test it live. Open ChatGPT or Claude in the meeting and ask it to compare one of their clients to two competitors. This is not a gimmick. Research on AI Overviews found they appear above organic results on more than half of real user queries, and that close to thirty percent of AI-cited domains never ranked on page one of traditional search. Your buyers are being handed summaries of you. If an agency's work cannot survive that compression, it is not finished work in 2026.

"What do you not do?""We do everything" means "we subcontract and mark it up." Good firms have edges and will tell you where theirs are.

Seventh, negotiate the things that actually go wrong

The disputes in branding engagements are boringly predictable. Handle them in the contract and you will avoid almost all of them.

Rounds of revision. Define what a round is. Vague revision language is where scope creep lives and where relationships die.

Who owns the files. Not just the final assets. The working files, the fonts, the licenses. Font licensing in particular catches people out, because the agency's license often does not transfer to you and nobody discovers this until launch week.

Named team commitment. If they sold you a specific creative director, that person should be named in the agreement with a minimum time commitment.

Kill fee and exit terms. Not because you plan to leave, but because agreeing how a breakup works while everyone is happy is much easier than agreeing later.

What "done" means. Write down the actual deliverable list. "Brand system" is not a deliverable. Twelve named artifacts is a deliverable.

Post-launch support. Rollout is when the questions start. Build in some hours or you will pay premium rates for them in a panic.

The red flags

Any one of these is a conversation. Two or more is a pass.

They agreed with everything you said. You are not paying for agreement. You are paying for someone with the standing to tell you that the thing you love is the thing holding you back.

The deck was gorgeous and the questions were shallow. Craft in the pitch, no curiosity about your business. This firm makes beautiful things that do not move numbers.

They cannot explain their own positioning. A branding firm that cannot articulate what it is and is not has failed the test it will charge you to pass.

They named your industry as the reason to hire them. Then the pitch was a portfolio of near-identical brands in that industry. You will get the next one.

Price came down fast when you hesitated. A firm that discounts 30% in one call was either overcharging or is about to under-resource you.

No process for after the launch. They are selling you an event. You need a system.

Everything is AI now, but they cannot say what changes. In 2026 everyone claims AI. Ask a specific question: what does AI change about how you build a brand system, concretely, and what does it not change? A real answer sounds like a method. A fake answer sounds like a press release.

The thing nobody tells you

You are not really choosing an agency.

You are choosing who gets to be in the room when your company decides what it is going to become. That is a strange and intimate thing to buy. The work is a byproduct.

The best branding partnerships I have been part of, on both sides, had one thing in common. Somebody said something in month two that was uncomfortable and correct, and the client did not flinch.

That is the whole test. Everything else is process.

So run the process properly, because it filters out the obvious mistakes. But make the final call on one question: which of these people will tell me the truth when it costs them something?

Hire that one.

If you want to find out whether that is us, our Brand Sprint is a small, paid, real piece of work. It is designed exactly for the paid diagnostic I described above, including the part where you get to see whether we will disagree with you.

Or tell us your problem and we will tell you honestly whether we are the right firm for it. Sometimes the answer is no. We would rather say it now than in month five.

And if you are still at the stage of building a list, here is which firms are built for which problem, us included, along with what we are not right for.

Frequently Asked Questions

How do I choose a branding agency?Diagnose your problem first: identity, positioning, brand architecture, or repositioning. These need different kinds of firm. Then name one internal decision-maker, write a two-page brief with your real business situation and your budget, shortlist three to five firms matched to your problem rather than your industry, and run a small paid diagnostic with your top two instead of asking for free spec work. Choose based on who will tell you the truth, not who presents best. For a breakdown of which firms are built for which problem, start there.

What questions should I ask a branding agency?Ask who specifically will do the work and what percentage of their time. Ask about a project that went badly. Ask what they talked a client out of. Ask for a case study with the business result and the baseline number. Ask what happens after they leave. Ask what they do not do. The goal is to get past rehearsed answers.

What are the red flags when hiring a branding agency?They agreed with everything you said. Beautiful deck, shallow questions. They cannot explain their own positioning. They pitched industry experience plus a portfolio of near-identical brands. They discounted heavily the moment you hesitated. They have no post-launch system. They claim AI capability but cannot say concretely what it changes.

Should I ask agencies for free creative work in the pitch?No. Spec work is produced by whoever is available, without research, to win a meeting rather than solve your problem. It teaches you little and it causes strong firms to decline. Run a small paid diagnostic instead: a workshop, audit or positioning sprint of two to four weeks with your top two or three. You will learn far more, including whether they can disagree with you productively. A Brand Sprint is one example of this format.

How long does choosing a branding partner take?Plan six to twelve weeks from brief to signed contract. Roughly two weeks to write the brief and align internally, two to three weeks to shortlist and meet, two to four weeks for paid diagnostics if you run them, and two weeks for contracting. Compressing this below a month usually means skipping the internal alignment, which is the part that determines whether the engagement succeeds.

What should be in a branding brief?Six things, in about two pages: the honest business situation, what must be true in two years stated as business goals, what is off the table, a narrowed audience of no more than three segments, your budget range, and your decision process and timeline. Long briefs assembled from multiple departments produce vague work.

Does the branding agency need to know my industry?Less than you think. Industry familiarity speeds up onboarding, but problem familiarity determines outcome. A firm that has repositioned complex technical businesses can usually reposition yours. A firm with a dozen brands in your exact category may hand you the thirteenth variation of the same idea, which is the opposite of what branding is for.

‍