Get a reality check on your startup idea.

Every startup idea needs a reality check. Get a 5-minute check before you build, commit, or spend real money.

Check Your Idea $29
The problem

Ask any AI if your idea is good and it will find a reason to say yes. Greenlight tells you what happened: to the market you are walking into, to founders like you, and to ideas like yours that came before.

How it works
The read

Four questions we answer before you commit or spend real money.

01
Is the market there, and can you reach it?
The Market Scan reads your market the way an investor would: what customers already pay for and what that says about how badly the problem is felt, whether the ground is moving toward your idea or away from it, and who controls access to your customer. Each read rated favorable, mixed, or hostile, with the evidence shown.
Step 1: Market Scan
02
Are you the right founder for this?
No other tool reads the founder. We read you, and your co-founders, against this specific market. Grounded in the frameworks investors use to judge teams, across five advisory dimensions: domain proximity, execution capability, channel and audience access, founder-customer empathy, and realistic execution risk. Honest where fit is thin.
Step 2: Founder-Market Fit
03
What happened to founders who tried ideas like yours?
Behind this question sits our record of 10,530 documented founder stories, every one tied to a public source. When your idea genuinely matches stories in the record, you see what actually happened to the founders who walked that path, and why. A chatbot can guess. A record can't be guessed.
Step 3: Hindsight Intelligence
04
Would top investors back you for this?
The best accelerators and VC firms have said, in public, what they look for and which mistakes they warn founders about. We built knowledge bases from that published thinking, for Y Combinator, Andreessen Horowitz and Sequoia Capital, and read you and your idea through it. Every claim cited to the video or essay it came from. Built from 1,756 videos and 1,414 articles, distilled into 19,275 sourced findings.
The Lenses

Five minutes from idea to read.

Get a Reality Check $29
The real reason

Every founder has heard "validate before you build." So why does almost no one do it?

Startups don't die because founders skip validation out of ignorance. They die because by the time validation would matter, the founder may no longer want the answer. Somewhere between the first spark and the first working version, curiosity hardens into commitment - and AI just made it happen in a weekend. We call it the Commitment Trap.

Read the full thesis →
Who Greenlight is for

Different kinds of founders. One avoidable mistake.

Each is about to make a costly, partially irreversible move. Each can avoid it with five minutes of evidence.

First-Time Founder
You have a full-time job. You earn well. You watch peers quit and build and raise, and you have an idea of your own. Your coworkers and partner think it sounds great. AI tools make building feel cheap. You're one weekend away from drafting your resignation.
"The decision you're about to make is worth six figures. Spend five minutes on evidence first."
The Serial Founder
You love building products. You've built before. Maybe even more than once. Months of work, a launch, and then the worst answer a market can give: silence. Not a no, not feedback, nothing. Now you have a new idea, and the old plan: build first, ask later. AI makes the next build even faster. Whether anyone wants it is still the question you skip every time.
"You've already paid for one silence. Five minutes of evidence before you pay for another."
Foreign Founder
You're building from outside the US. Cheaper, faster, with talent you trust. You believe your product fits the US market. Maybe you're already planning the flight. But the US market doesn't behave like your home market, and you haven't asked it anything yet.
"Don't move countries before you have a greenlight from the one you're moving to."
Get a Reality Check $29
The other reason

Your AI was trained to agree with you. It even has a name: sycophancy.

Ask a chatbot whether your idea is good and it will find a reason to say yes, not because it is careless but because it was rewarded for it. The models are trained on human feedback, and people rate agreement highly. Anthropic's researchers measured it; OpenAI rolled back a model for it. For a founder, that is the most dangerous kind of encouragement there is: an evaluation optimised to make you feel good about the thing you are already committed to.

That is why a Reality Check is never asked whether your idea is good. It is asked what the real data shows, and the response is analyzed from fifteen different angles and lenses, with every claim traceable to its source.

Read the sycophancy thesis →
Which one sounds like you?

The traps that look like progress.

Six patterns of validation that aren't validation. Each one feels productive in the moment. None of them are the market.

01
The AI-Validated Founder
You asked a chatbot if your idea was good. It gave an enthusiastic yes.
"A chatbot will never tell you your idea is bad."
02
The Inner Circle Prisoner
Your family is supportive. Your friends are encouraging. Or one skeptical voice killed an idea that might have worked.
"Your friends and family are not the market."
03
The AI-Empowered Builder
AI made building ten times faster. So you ship before you ask. You pivot based on one customer's reaction, then the next.
"Shipping is not asking."
04
The Product Perfectionist
You've improved the product six times and traction never came. You assume the product is the problem.
"The market didn't reject your product. It never knew it existed."
05
The Idea Collector
You generate ideas constantly. You analyze them, refine them, and talk about them. You never test one, because the next idea is always more interesting.
"Stop collecting ideas. Start testing them."
06
The Serial Pivoter
You've pivoted three times this year. Each pivot felt like learning. But every pivot was based on a new unvalidated assumption.
"You're not pivoting. You're guessing twice."

Your startup idea. Checked in five minutes.

Get a Reality Check $29
What we believe

We don't follow the playbook. We follow the evidence.

The playbook says build first and find out later. These six beliefs say otherwise, and every one of them shapes how we read your idea.

01
Cheap building is a trap, not a gift. AI lowered the cost of building. It did not lower the cost of building the wrong thing.
02
Emotional commitment is the most expensive thing a founder owns. Once you're locked in, you can't hear the market when it says no.
03
More than 10,000 founders already ran versions of your experiment. Reading their outcomes takes five minutes. Repeating them takes years.
04
Evidence, not a verdict. We show you what we found and where it came from. The decision stays yours, which is where it belongs.
05
The founder is half the answer. The same idea lives or dies depending on who runs it. That is why we read you, not just your market.
06
Flattery is free everywhere. An honest read is worth paying for. We would rather lose your $29 than tell you what you want to hear.
For Incubators & Accelerators

Incubators & Accelerators
get a structured signal
before the cohort starts.

Every program admits on instinct. Greenlight puts a Reality Check on every applicant: the market, the founder, and the record of ideas like theirs, before the cohort begins. Optionally read through your program's own methodology, kept confidential to you.

If you're running an incubator or accelerator program and want structured signal on your applicants, here's how it works.
See the program offering →
Get in touch

Questions, partnerships, or just curious?

Reach out if you're an incubator, accelerator, or VC firm exploring partnership opportunities, or if you just want to say hello. We'd love to hear from you.

Thank you. The Greenlight AI team will reply to you shortly.