I ran an intensive earlier this month with a food and beverage agency, about five years old, eight core people plus contractors, doing $40K to $50K a month with a peak month around $71K. Twelve or thirteen clients, all of it inbound off referrals and SEO, and they rank first for fifteen to twenty food and beverage search terms.
On paper that’s a niched agency, and it’s the case study you’d put in a niching post.
Underneath it, they were serving different ICPs with completely different problems, client revenue was scattered all over the spectrum, and the community and social work couldn’t be tied to revenue, so it churned. And there was no clear promised outcome anywhere in the business, which is the piece that mattered most, because without one there’s nothing for the marketing to say. The founder was also account manager on every single account, which caps the whole thing at his attention.
Niche chaos is what you get when you’ve chosen four times and retired none of them. The agency has a niche on the website, a different one in the pipeline, a third one paying the bills, and a fourth one nobody’s admitted is dead yet, and because each one carries a different buyer with a different problem, the agency can’t promise a specific outcome to any of them.
Quick Take
- Niche chaos looks like focus from the outside. Ranking for your category’s terms and turning down nothing are compatible states, and that’s what makes it hard to see from inside.
- The line between deliberate multi-industry and chaos is whether it was chosen. An agency serving four industries on purpose can tell you why each one is there and what it’s worth.
- Run the cut on the problem first, not on the industry. The more painful the problem, the more likely somebody pays to make it stop.
- Then ask which claim you can defend: experience, understanding, or resolve. If the honest answer is none of them, you don’t own that niche, you just want it.
- The cost of staying runs deeper than unfocused marketing, because you lose the promised outcome, so every proposal gets built from scratch and every price turns into a negotiation.
What this covers
- What niche chaos is, and why it hides so well
- Serving several industries on purpose vs. niche chaos
- How to run the cut when you’ve got four candidates
- The three claims, and which one you can defend
- What staying in chaos costs you
- The math that makes a small niche look big
What is niche chaos?
Niche chaos is the state of having chosen a niche more than once without ever retiring the previous choice. Each one arrived for a reason that made sense that quarter, none of them got closed out, and now they coexist, splitting the founder’s attention and the agency’s story between buyers who have nothing in common.
The reason it hides is that the surface metrics of a niched agency and a chaotic one look identical. The food and beverage agency ranked first for fifteen to twenty terms in their category, ran fully inbound, and had a category word in their positioning. Everything you’d check from outside came back clean. What you couldn’t see from outside was that the clients underneath that category had different problems, different budgets, and no shared definition of what a win looked like.
So the diagnostic isn’t “do you have a niche.” Almost everybody has a niche in the sense of a word they say. The diagnostic is whether the clients inside it share a problem, and whether you promise them all the same outcome. When the answer to either is no, you have a category you market in rather than a niche you serve, and those behave very differently once you try to price the work.
Is serving several industries on purpose the same as niche chaos?
No, and the line between them is whether it was chosen. Niching down was never about saying no to all other business, and it’s about the work being a deliberate choice instead of accepting everybody by default.
An agency serving four industries on purpose can tell you why each one is there, what it’s worth to them, and what happens to it next. An agency in chaos took four and never retired three, and each one arrived because somebody said yes to revenue during a thin month. Same client list, completely different business.
The practical version of “chosen” is lead scoring. Your niche is the perfect score, and other criteria can still qualify a lead, so a strong-fit lead outside the industry is still a far better client than a junk lead you only took for the money. That’s a system making the call, and chaos is the absence of one.
What tips an agency over is almost always a thin quarter. You take the client who doesn’t fit, it works out fine, they’re pleasant, they pay on time, and eighteen months later it’s a service line nobody ever decided to have. The tell is that you can’t name the meeting where you decided to serve them.
How do you pick a niche when you’ve got four?
Start with the problem, not the industry. There has to be a clear problem that needs solving, and the more painful that problem is, the more likely somebody’s going to want a solution for it badly enough to pay. An industry with no urgent problem is a demographic, and demographics don’t buy anything.
Before you go further, run the math on whether the problem is even worth owning. Is it painful enough that people pay to make it stop, does it recur, is it big enough to fund a real budget, is it solvable, and are the people who have it reachable? Anything that fails one of those gets killed before you spend another minute on it.
Then the skills question, which is mostly a moral one. You should have the skills to solve this problem, and if you can’t solve it, don’t touch it. That’s an ethics filter rather than a strategy one, and it runs before the strategy ones do.
The last question is the interesting one, and it’s the one that decides between your remaining candidates.
Which claim can you defend: experience, understanding, or resolve?
Once a problem passes the math and the skills check, what’s left is whether you have a credible claim to it, and there are three ways to have one. Which of the three you can defend in a room of practitioners is the real test, and if the honest answer is none of them, you don’t own that niche, you just want it.
The first claim is experience, which means you have proof you’ve solved this before and you can describe the buyer’s Tuesday without being briefed on it. It’s the strongest of the three and the slowest to build, because there’s no shortcut to having done the thing.
The second is understanding, and this is the one agencies underrate. You’ve worked in a different space solving a similar problem, you’ve seen enough versions of it to know where it breaks next, and you’re bringing that across into a new area. You don’t have direct experience in this niche, and you have a real read on the mechanics of the problem, which is often worth more than a client logo.
The third is resolve, which means your reason for being here runs deeper than the invoice. You have enough conviction about solving this problem that people come in on the strength of watching you care about it. It’s the least tangible claim and it’s genuinely a claim, because buyers can tell the difference between someone who picked a problem off a spreadsheet and someone who belongs to it.
Think of the three as circles that resize depending on who you’re selling to, since the buyer decides which one has to be biggest. Price moderates it too: low-stakes work lets resolve carry the weight on its own, and high-stakes work demands experience or understanding stacked behind the conviction, because nobody hands you a bet-the-company project on enthusiasm.
Which is why the cut usually resolves cleanly once you run it honestly. The common hard case is a founder with deep experience in a niche that has no urgent problem, up against a niche with a screaming problem where they’ve got one project’s worth of history. The problem test kills the first one before the claim test even runs, and the second survives on understanding rather than experience, which is a legitimate claim as long as you’re honest that it’s the one you’re making.
What does staying in chaos cost?
More than unfocused marketing, and the second-order costs are the ones that surprise people.
You lose the promised outcome, and with it the ability to price. Serving four niches with four different problems means you can’t promise a specific result to anybody, so every proposal gets built from scratch and every price becomes a negotiation instead of a statement. That’s where the margin goes, and most owners read it as a sales problem.
The second cost is capacity, and it’s the one nobody sees coming. Every extra niche is another context the founder has to hold in their head, so the founder ends up account manager on everything, because nobody else on the team can carry four different sets of client problems at once. The food and beverage agency hit exactly this. That’s how an owner ends up capped without ever having made a bad hire.
Third, nothing compounds. Four niches means every project is close to a snowflake, so senior people spend their time on work that should be junior, and none of the process you build for one client transfers to the next. You do the same amount of work every year and get no cheaper at any of it.
The math that makes a small niche look big
The hardest cut I’ve run was with an SEO agency that wanted to take their marketing nationwide. They were already dominating locally, ranking number one in their city for SEO, and the pull toward a bigger map is strong when you’re winning a small one.
So we did the math instead of arguing about it. We pulled Chamber of Commerce data for recreation companies inside their own city, assumed only 5% of them would be in market at any given time, and ran the ceiling. Even on those assumptions, working nothing but recreation companies in one city, the business could get to $6 million.
Rolling the math back and showing them the actual number is what made it obvious. They put their heads down, doubled down on the focus, and grew quickly from there. Nobody talked them out of going national with a philosophy about focus, because philosophies don’t move founders, and a ceiling they could see on a spreadsheet did.
This is the answer to the fear underneath most niche chaos, which is that choosing will cost you all the other revenue and leave you looking like a one-trick shop. Run the ceiling on the narrowest version of your niche before you assume it’s too small, and roll the choice out over time with lead scoring rather than firing clients on a Friday. Most owners have never run that number, and it’s almost always bigger than they think.
Worth noting that the clients I’ve watched niche successfully mostly did it after we started working together rather than before. One went to manufacturing, one to CPG, one to family law, one to vertical SaaS. None of them arrived knowing, and all of them found it in their own history rather than on a market map, which is the real move: look at your past engagements for the niche that’s already authentically yours instead of the one that looks biggest on paper.
The Monday version
List every industry you’ve invoiced in the last twelve months. Next to each one, write the meeting where you decided to serve them. The ones with no meeting are the chaos, and that list is usually longer than an owner expects.
Then take your two strongest candidates and run them through in order: is the problem painful, recurring, funded, solvable, and are those people reachable, do you have the skills to solve it, and which of experience, understanding, or resolve can you defend if a practitioner pushes back. A candidate that survives all three is your niche. A candidate that survives on resolve alone is workable for lower-stakes offers and won’t hold up on the big ones.
If you’re an agency owner who can name four niches and can’t name which one is yours, this is the work we do inside the Dynamic Agency Community.
FAQ
What is niche chaos?
Niche chaos is having chosen a niche several times without retiring any of the earlier choices, so the agency ends up with one niche on the website, another in the pipeline, a third paying the bills, and a fourth nobody’s admitted is dead. Because each one comes with a different buyer and a different problem, the agency can’t promise a specific outcome to any of them, and that’s what breaks the pricing.
Can an agency serve multiple industries without being in niche chaos?
Yes, as long as each one was chosen. The test is whether you can say why each industry is there and what it’s worth to you. Lead scoring is how this works in practice: your niche is the perfect score, other criteria still qualify a lead, and a strong-fit prospect outside the industry beats a junk lead you took in a slow month.
How do I pick a niche for my agency?
Run three filters in order. Is there a clear, painful, recurring, funded, solvable problem and can you reach the people who have it. Do you have the skills to solve it, which is an ethics question before it’s a strategy one. And which claim can you defend, experience, understanding, or resolve. Most candidates die at the first filter, which is why starting with the industry instead of the problem wastes so much time.
What if my niche is too small?
Run the ceiling before you assume it. One agency I worked with wanted to go national until we pulled Chamber of Commerce data for a single vertical in their own city, assumed only 5% were in market, and found a $6 million business sitting inside the boundary they were trying to escape. Most owners have never run that math on the narrow version.
Do I have to fire my non-niche clients?
No. Claiming a niche makes other work a deliberate choice rather than a default yes, and the rollout happens over time through lead scoring rather than through a mass exit. Firing revenue on a Friday is how owners talk themselves out of niching entirely.