How to choose a niche narrow enough to win

Bizer · 2026-10-01

A niche is not a topic. "Fitness" is a topic. "Bookkeeping" is a topic. A niche is a specific group of people with a specific problem, in a place where you can reach them, who already spend money on that problem. If you cannot picture the person, you do not have a niche yet.

Most owners pick one too broad. They worry that narrowing will shrink the market, so they describe themselves to everyone, and nobody in particular recognises themselves in the description.

What does a good niche look like?

Compare two businesses. One is "a bookkeeping service for small businesses." The other is "bookkeeping for commercial fishing and shrimping operations in Terrebonne and Lafourche parishes, built around the season."

The second one is smaller on paper. It is also the one that wins. It knows its customers' busy months, it understands fuel and ice as cost lines, it can name the docks and the associations where those owners gather, and every one of those owners who hears about it thinks "that is for me." The first one competes on price with every bookkeeper in the state and with software.

A niche worth having passes four tests.

1. You can list 50 of them by name, or at least find 50 in an afternoon. If you cannot find them, your marketing cannot either.
2. You can reach them cheaply. A trade association, a Facebook group, a supply house counter, a trade show, a newsletter they all read. One shared place is enough.
3. They already pay to solve the problem, with money, with time, or with a worse competitor. A problem nobody spends on is a problem nobody will pay you for.
4. You have an edge. You worked in the industry, you speak the language, you know the regulations, you have done the job for ten of them already.

The fourth test is the one people skip, and it decides the price you can charge.

How do you test one in a week?

Talk to them. Pick a candidate niche and have five conversations with people in it before the week is out. Ask what the problem cost them last time, who they use now, and what annoys them about that person. Do not pitch.

You are listening for three signals: the same complaint from different people, a dollar figure or hours lost that they mention without being asked, and someone asking when you can start. Two out of three in five conversations is a niche worth pursuing. None out of three means try the next candidate, not the next script. The questions that get honest answers take some practice, because the natural ones invite polite lies.

Then count them. The Census Bureau's County Business Patterns data reports the number of business establishments by industry code and county, free. If your niche is other businesses, it tells you whether there are 40 of them in your parish or 4,000, and whether 40 is enough.

Do the arithmetic. If an average customer is worth $3,000 a year to you and you need $90,000 in revenue, you need 30 customers. Thirty out of 400 reachable prospects is a plan. Thirty out of 45 is a stretch, and you will want a second niche later.

Is it possible to go too narrow?

Yes, but it is rarer than people fear, and easy to spot. You are too narrow when the total number of possible customers cannot reach your revenue target even at a good share of the market, or when the niche depends on one employer, one contract, or one regulation that could vanish.

You are not too narrow just because the description sounds small. Narrow describes who you market to. It does not cap who you serve. The fishing bookkeeper will pick up a marina, a seafood processor and a boat repair shop through referrals without changing a word of the website. If the numbers come out close, go narrow or rethink walks through that decision.

What does niching down cost you?

You will say no to work. A lead outside the niche calls, and taking the job pulls you away from the people you are trying to become known to. Some owners handle this with a referral partner. Some take the job quietly and never market to that kind of customer. Either is fine. What fails is rebuilding the business around whoever called last.

It also costs some optionality. A niche is a public commitment. Changing it after a year means new words, new places to show up and, sometimes, a new name.

Our view: pay that cost. A business known for one thing to a few hundred people beats a business vaguely known to everyone, nearly every time, because referrals only travel when the person passing your name on can say exactly who you are for.

What is uncertain

No test tells you a niche will work. Five conversations can mislead you if all five people are friends of a friend, and Census counts lag by a couple of years. The only real proof is a stranger in the niche paying full price. Get that before you print anything.