An idea you read in a listicle is an idea a thousand people read the same week. Four sources that produce ideas attached to someone you can name.
Nearly every article on how to find SaaS ideas is itself a list of ideas. Lists are the wrong format, because an idea you read is an idea a thousand other people read the same week — and none of you have any advantage in executing it.
All four produce ideas attached to a specific person you can name. That attachment is the whole point.
The test for a real idea is not whether it sounds good. It is whether you can name one person who has the problem and say how you would reach them this week.
Whatever job you have held, you have watched people waste time on something. That knowledge is the most durable idea source available, because it is the one thing a competitor cannot copy in a weekend.
Three prompts that produce specific candidates:
The third is the strongest, because it comes with a budget attached. You already know the problem is worth money to someone, which clears the hardest validation gate before you start. For a deeper pass at finding an idea in your own industry specifically, the dedicated guide walks through the same prompts in more depth.
Review sites for existing software are the most underused idea source on the internet. A one-star review is a person with budget, an active problem and a documented reason for being unhappy.
Look for the same complaint repeating across many reviews of the same product. One angry customer is noise. Forty people describing the same missing capability is a specification written by the market.
The idea that comes from this source is rarely a full competitor. It is usually the narrow thing the incumbent does badly, sold to the segment the incumbent does not care about.
Watch how small businesses actually operate and you will see the same patterns everywhere: a spreadsheet acting as a database, copy-paste between two systems, a checklist in a notes app, a person retyping information from one screen into another.
Each of those is a product that has not been built for that specific niche yet. Horizontal versions almost always exist; the opportunity is the vertical version that fits one industry's workflow exactly.
This source rewards proximity. You have to see the process, which means talking to people or having worked near them — customer interviews are how you get access without an audience.
A new regulation, a platform policy change, an API deprecation, a tax rule. Each creates work that did not exist before, for people who were not looking for software last month.
Timing is the advantage and the risk. Early enough and you own the category; too early and nobody is looking yet; too late and three competitors are funded. The window is usually months rather than years.
Sources produce candidates. Most should be discarded within an hour, and three questions do it.
| Question | Discard if |
|---|---|
| Can I name one person with this problem? | You can only describe a demographic |
| Where do these people gather? | You cannot name a specific place |
| Is anyone paying to solve it today? | No spend anywhere, in any form |
Ideas that survive all three go into a shortlist. Two survivors is enough to move on — see how to choose between two SaaS ideas for the tie-breakers.
Keep a running note of candidates and revisit it monthly rather than trying to generate ideas on demand. Ideas found while looking for one are consistently worse than ideas noticed while doing something else.
Three reasons, and they compound.
No distribution. An idea from a list arrives with no audience attached. You would start from zero on the hardest part of the business.
No domain advantage. You would be building for people whose work you do not understand, competing with someone who does.
No filter. List items are chosen because they sound plausible in a paragraph, which is uncorrelated with whether anyone will pay.
The exception is using a list as a prompt rather than a menu — reading twenty ideas to notice which one connects to something you already know. That is idea generation using your own knowledge, with the list as a trigger.
Do not build. Test it against the nine-test validation checklist, which needs no product and takes two to three weeks.
If the idea passes, positioning is the next decision and it sets your pricing ceiling — the positioning framework starts by naming the alternative you replace.
If you already have a working product and no business around it, the idea stage is behind you — the micro-SaaS guide and the full validation process cover what comes next.
Use four sources: work you have already done, repeated complaints about tools people already pay for, manual processes you can observe in small businesses, and regulatory or platform changes that force new work. All four produce ideas attached to a person you can name.
An idea from a list comes with no distribution, no domain advantage and no filter. A thousand other people read it the same week, and list items are selected for sounding plausible in a paragraph rather than for anyone being willing to pay.
Work you have already done. Domain knowledge is the one advantage a competitor cannot copy in a weekend, and the strongest version is anything your employer paid an agency or contractor to do — the budget is already proven.
Three fast filters: can you name one specific person with the problem, can you name a specific place those people gather, and is anyone already paying to solve it in any form. Discard candidates that fail any of the three within an hour.
Two is enough. More than that turns selection into a research project. Keep a running note of candidates and revisit monthly, because ideas noticed while doing other work are consistently better than ideas generated on demand.
Tell Marcus the idea and where it came from. You get the three filters applied and a straight answer on whether to continue.
Try GhostCoach free →14-day free trial · cancel anytime · 30-day money-back on Lifetime