The niche scorecard: how we rank 42 markets against your background

Picking a niche is the single most expensive decision in an AI agent business. Get it right and the build compounds. Get it wrong and you burn six months learning a market you have no business being in.

We score 42 niches against every operator who comes through. Below is how the scorecard actually works, what the inputs are, and the three dimensions that matter most.

The framing question

We do not ask operators “what niche do you want to be in?” That question puts the burden on you to already know an answer that is almost impossible to know from the outside. Instead we ask: where do you have an edge that we can amplify?

Edge usually shows up in one of four ways:

• Background. You’ve operated in this industry, sold to it, or built products for it.
• Network. You have warm contacts who are buyers or who can introduce you to buyers.
• Affinity. You understand the buyer psychology, the vocabulary, the rituals, the politics.
• Asymmetric data. You know something the rest of the market does not know yet.

If you have any one of these, we can usually build around it. If you have none, the build is a gamble. The scorecard exists to surface which niche you have the most edge in across the 42 we serve.

The three scoring dimensions

Every niche gets scored against three dimensions, in this order.

1. Market gravity

Does the niche have enough demand that a single operator can absorb meaningful revenue without saturating the market? We look at three sub-factors:

• Total addressable market in the operator’s geography (or globally, depending on the offer).
• Buyer density — how clustered the buyers are. High density means lower acquisition cost.
• AI receptivity — how much the buyer pool already expects, accepts, or actively wants AI in their workflow.

A high-gravity niche has 10,000+ qualified buyers, clustered around 3 to 5 cities or platforms, and a buyer pool that is already evaluating AI tools. Healthcare, real estate, and bookkeeping all score high here. Restaurants and individual creators score lower.

2. Operator fit

The match between the niche and the operator’s edge. We look at:

• Domain background. Years operated in or adjacent to the niche.
• Warm network. Can you list 20 named buyers in the niche right now?
• Sales motion fit. Does the niche close via calls, demos, referrals, or self-serve? Match to operator preference.
• Capital tolerance. Some niches require months of warmup. Others close inside two weeks.

Operator fit is the difference between “this niche is winnable” and “this niche is winnable by you.” High fit means you walk into the first sales call already knowing the buyer’s language. Low fit means you spend the first six months learning vocabulary you should have brought to the table.

3. Acquisition economics

The unit economics specific to the niche. We look at:

• Cost per qualified lead from our typical channel mix in that niche.
• Conversion rate from booked call to closed customer (averaged across our active operators).
• LTV / CAC with the specific offer we’d build.
• Payback period. How fast does the operator recoup the build cost?

Acquisition economics are the dimension operators most often skip when they pick a niche themselves. A high-margin niche with terrible CAC is a worse bet than a moderate-margin niche where every lead is essentially free.

What the scorecard output looks like

Every operator who books a strategy call gets a personalized scorecard at the end. It looks something like this:

• Top 3 niches. Ranked by composite score, with the dimensions broken out. We tell you exactly why each made the top three.
• Why-not list. Niches that scored high on gravity but low on fit, with the specific reason.
• The build. What we would build first if you picked the top niche. Specific agent, specific offer, specific acquisition stack.

If we can’t get to a top three with a strong composite score, we tell you on the call. Not every operator is a fit for every niche, and not every operator is a fit for our model. We’d rather tell you on the call than pretend.

Three rules that govern niche lock-in

Three things we do that other agencies usually don’t:

We do not lock until we both agree. The niche is your business for the next 18+ months. The scorecard is a recommendation, not a verdict. If you see something we missed about your background, we re-score.

We do not lock you in. If the niche dies after launch (market shifts, regulation, etc.), we move you to a new one at no extra charge. Niches do die occasionally. We’re built to handle it.

We say no. If we cannot get to a strong top three for you, we’ll tell you we’re not the right fit. Better to lose the engagement than to ship a build into a niche neither of us believes in.

Where to start

The scorecard runs live on the strategy call. We bring your background, the scorecard methodology, and the 42-niche dataset. You bring the operator context only you have. We get to a recommendation in 30 minutes.

Book a strategy call. If we agree on a top niche, we walk you through what we’d build. If we don’t, you leave with a clearer view of your own positioning, which is worth the 30 minutes either way.