Yes. Generating ideas is free. You only need a free SeedAngels account (email or Google, no card) so your ideas are saved to your history and you can come back to them.
Free Startup Idea GeneratorBuilt on Y Combinator's method for finding ideas
Most generators hand you a list anyone could have obtained. This one starts from your skills and your experience, then scores each idea against the questions YC asks.
- 5 ideas grounded in your own background
- Each one scored out of 10, tarpits flagged
- Pick one and validate it in a click
- Free
- Confidential
- Based on the YC method
Tell us what you bring
The more concrete you are, the more specific your ideas will be. Be factual about your skills.
Your skills, technical or otherwise. This is what makes an idea yours rather than anyone else's.
0 / 2000
Past employers, industries, internships, and the processes you watched waste everyone's time.
0 / 2000
Things that annoy you often enough that you'd pay to make them go away.
0 / 2000
New technology, regulation or behaviour that makes something possible now that wasn't before.
0 / 2000
Fill in the first two fields: they are what founder-market fit is judged on.
What you write stays confidential and is used only to generate your ideas. A free account is needed to see them, so they are saved to your history.
How this free startup idea generator works
No brainstorming, no generic list. Your background goes in, a ranked shortlist comes out.
- 01
Tell us what you bring
Four short fields: what you are unusually good at, where you have worked and what was broken there, and, if you like, the problems you have felt and the recent changes you have noticed.
- 02
Get five ranked ideas
Within a minute: five ideas, each built on something you told us, scored out of 10 and ranked. The best one opens in full. Not convinced? Ask for different ideas from the same profile, or edit it and run again.
- 03
Pick one and validate it
One click carries the idea you pick into the startup idea validator, with its problem and solution already filled in. Every generation is saved to your history, so you can come back to the others.
What you get for each idea
Each idea comes back assessed against the questions YC actually asks, not merely described.
E-invoice reconciliation for accounting firms
From a broken industryA platform that automatically matches mandated e-invoices, bank transactions and accounting entries for the SME clients of French accounting firms.
Score: 8 out of 10
- Why you
- One founder has built payment infrastructure and open-banking APIs; the other spent 10 years auditing SME accounts and watched reconciliation done by hand at every month-end.
- Why now
- E-invoicing becomes mandatory for French businesses from September 2026, creating structured invoice data and a deadline every firm must meet.
- Market size (estimate)
- France: roughly €300M–€800M a year across SME accounting, invoicing and reconciliation software.
Already on this ground
- Pennylane
- Cegid
- Sage
- Excel, email and bank portals
Your first step this week
Interview 10 accounting-firm partners about their last month-end close: collect the unmatched invoices and the time spent resolving each one.
- 01The problemFramed as a problem, never as a product in search of a use.
- 02Who has itThe precise customer, not a vague market.
- 03Why youFounder-market fit, anchored in what you told us.
- 04Why nowThe recent change that makes this possible.
- 05Market sizeAn honest estimate with its geography, no false precision.
- 06Problem intensityHow much it hurts those who live with it.
- 07Already on this groundReal competitors and the status quo, such as a spreadsheet and phone calls. Never an invented name.
- 08Tarpit warningFlagged when the idea resembles those many teams have already failed on, with the obstacle named.
- 09Biggest riskThe real reason it could fail, not a polite caveat.
- 10Your first stepSomething concrete to do this week, usually talking to the right people about what they actually did.
How to read the score out of 10
The score weighs four things: why you, how much the problem hurts, why now, and whether it is a tarpit. The ideas are ranked by it.
- Rare9–10
Obvious founder fit, an acute problem, a credible timing and no tarpit.
- Strong7–8
Solid on three of those four. Worth validating first.
- Interesting5–6
One serious weakness: a weak fit, a lukewarm problem or a narrow market.
- Weak1–4
Weak on several points. A tarpit idea never scores above 5.
Where good startup ideas actually come from
The best ideas are rarely born of a brainstorm. YC partner Jared Friedman sets out the "recipes" that produce them, and they all start from something you already have. This generator applies four of them, hence the questions about you.
- 01
Start from your unfair advantage
What can you, or your team, do that most people cannot? An idea founded on real expertise cannot be copied on a simple decision.
- 02
Mine what you have lived
Your past roles, the industries you know, the problems you would like to see solved. You already know things about them that no amount of research can replace.
- 03
Ride a recent change
A technology, a regulation, a shift in habits. What has just become possible or necessary opens a gap the incumbents did not have.
- 04
Stay out of the tarpits
Some ideas look obvious because dozens of teams have tried them and failed for a structural reason. When an idea resembles one, we say so and name the obstacle.
Nothing about you is made up
If you leave the problems field empty, no idea claims to come from a problem you have felt. Competitors are real companies or what customers do today, never invented names. Market sizes are labelled as estimates, and the idea you pick is checked against live web sources in the validator.
The talk this tool is based on
His “recipes” became the questions we ask you. His evaluation criteria became the score on each idea you receive. Worth its 45 minutes.
How to Get and Evaluate Startup Ideas, by Jared Friedman (Y Combinator Startup School). Watch on YouTube
Frequently asked questions
The framework set out in "How to Get and Evaluate Startup Ideas", the YC Startup School talk by Jared Friedman, embedded above. His “recipes” became the questions in the form; his criteria, the score on each idea. We are not affiliated with Y Combinator, we simply think it is the best public material on the subject.
This tool is the step before it. Use the generator when you want to start a company without yet knowing what to build; use the startup idea validator when you already have an idea and want to know whether it stands up. Most founders use them in that order, and in the app a single click takes the idea you pick from one to the other.
Ask for different ideas: the generator keeps your profile and steers away from the ideas you have already seen. You can also edit your profile, usually by adding a recent change or a problem you have felt, and run it again. Every generation stays in your history.
A tarpit idea presents a very widespread problem and an apparently obvious solution that dozens of teams have already attempted without success, for a structural reason invisible from the outside. Typically, apps for organising outings with friends. They are dangerous precisely because they seem promising. When a generated idea resembles one, we say so, name the obstacle, and its score cannot go above 5 out of 10.
Because founder-market fit is one of the first things investors look at, and because an idea anyone can execute will be executed by anyone. Starting from a theme produces ideas in which you have no particular advantage. Starting from what you know produces a shorter list, but one you are genuinely placed to win.
No, rather the opposite. Competitors prove that someone is willing to pay to have the problem solved. Empty ground more often means there is no market than that you have found an untouched one. That is why each idea comes with the players already there, including the status quo.
Treat them as a first estimate. They come from the AI's general knowledge, are labelled as estimates, and real company names are only given when the AI is confident they exist. Before you commit, run the idea you pick through the validator, which checks the market against current web sources.
Your profile is saved with your ideas in your private history, so you can edit it and generate again, and you can delete any generation. It is never published, sold or used to train anything.
Found the one? Put it to the test
A good idea on paper is a starting point. From the idea you pick, SeedAngels takes you through the next steps in the same place.
- Validation: a few questions about real facts and a score out of 100
- Web research on your market, with sources you can check
- A Lean Canvas, a financial forecast and a business plan built from it
- A pitch deck ready to show investors