Early opportunity usually looks unattractive
By the time an opportunity looks obvious, the easy version of it is gone. Obvious opportunities attract capital, attention and competition in that order, and the window closes fastest in exactly the categories everyone can see.
Early opportunity has a signature instead: it is small, unglamorous, and it is being discussed by people who are not trying to build anything. The conversation is happening in a trade forum, a service business's back office, or a thread about a spreadsheet that someone maintains by hand every month.
That is why the practical skill is not idea generation. It is evaluation. Anyone can produce twenty ideas in an hour. The scarce ability is to look at one and judge whether it is early, or simply bad.
Three places early signal actually lives
- Complaint density. Not one complaint, but the same complaint repeated almost verbatim in different places by people who do not know each other. Identical wording across unrelated forums means the problem is structural, not personal.
- Manual workarounds. A monthly spreadsheet, a copy-paste ritual, a VA doing something tedious at 6am. Every workaround is a budget already being spent, badly, and a buyer who has already admitted the problem exists.
- A deadline that forces behaviour. A regulation, a platform change, a tax rule, an expiring contract. Something that makes the status quo stop working on a known date. Timing is not a vibe; it is a date on a calendar.
None of this requires privileged data. It requires reading where the operators are, rather than where the commentary is.
Signal, pattern, opportunity
Three words, in a fixed order, and most people skip the middle one.
A signal is one observation. A complaint, a workaround, a new rule. A pattern is the same signal recurring across independent sources, with a cost attached. An opportunity is a pattern where you can identify the buyer, the budget, and how you would reach them.
The gap between pattern and opportunity is where most ideas die, and it is usually a distribution problem rather than a product one. If you cannot name the first twenty people who would pay, and how you would find them, you have found a problem worth writing about and not yet a business.
A seven-factor score, and what each factor is really asking
Scores are not predictions. They are a way to make your judgement explicit so it can be argued with, including by you. These seven run to a hundred points.
- Demand (20). Is anyone already spending money to solve this badly? Evidence of spend beats evidence of interest every time.
- Pain (20). How much does the current situation cost the buyer? In hours, in money, in risk. Quantify it or it is not a factor.
- Monetization (15). Can you charge for this without changing how the buyer already behaves? Products that require a new purchasing habit are products that need two wins.
- Timing (15). What changed recently that makes this possible now? If the answer is nothing, you are early or you are late.
- Competition (10). High competition is evidence of a real market and a warning about your margin. The question is whether the incumbents are serving this buyer well.
- Market size (10). How many buyers are there, and how often does each one buy? A small number of frequent buyers usually beats a large number of one-time ones.
- Founder fit (10). Can you get in front of these buyers, and will you still care in two years?
The weights matter less than writing each one down with a number. The moment a factor has a number beside it, you can see which assumption is carrying the whole score.
Founder fit is the factor you cannot fake
A seventy-eight that you are wrong for loses to a sixty-four that you are right for. Not because the smaller opportunity is better, but because the score assumes execution, and execution is the constraint that actually binds.
Fit is not passion. It is proximity. Do you understand the buyer's language, do you have a way in, and can you survive the boring middle of the problem? Someone who has worked inside a dental practice will beat a better-funded outsider on dental software every time, because they know which of the twenty obvious features nobody would pay for.
Write the anti-thesis first
Before you spend a week on an idea, write the three strongest reasons it does not work. Not the three you can dismiss. The three that would make you abandon it if a friend said them to you.
This is not pessimism, it is calibration. Every idea you are excited about arrives with a story already attached, and the story is the thing you end up testing rather than the idea. Writing the anti-thesis first means your validation work has somewhere to go: it either kills the idea cheaply or gives you the sentence you will later use in a pitch, because you already know the objection.
A worked example
Say you find a pattern in scheduling for small trade businesses: the same complaint about quoting and booking, repeated across three forums over two years, with owners describing a spreadsheet and a phone.
- Demand 85. Owners already pay for a booking tool or an answering service, and complain about both.
- Pain 80. A missed call is a lost job; the cost is concrete and repeated weekly.
- Monetization 70. They already buy software monthly, so no new purchasing habit is needed.
- Timing 60. Nothing has changed recently. That is a warning, not a green light.
- Competition 55. Several tools exist; most are aimed at larger crews.
- Market 70. Many buyers, and each one buys for years if the tool sticks.
- Fit 40. You have never worked in the trade and have no way into the buyer's world.
The weighted score is 69. That looks like a pass, and it is not. Demand and pain are strong, but timing is flat and fit is weak, which means the two factors most likely to decide whether this works are the two you cannot currently fix. The honest next move is not to build a prototype. It is to test whether you can get twenty owners on the phone, because that answer decides everything else.
That is what scoring is for. Not to rank ideas against each other, but to show you which assumption you are standing on.