Five questions about evidence, twenty points each, one score out of a hundred. Market, buyer, pain, differentiation and willingness to pay. The scoring is deliberately unkind: a yes requires evidence from outside your own head, which is why most first attempts land between forty and sixty and why that is useful rather than discouraging.
Score = sum of 5 answers, where yes = 20, maybe = 10, not yet = 0
Five yes/no/maybe questions across Market, Buyer, Pain, Differentiation, and Willingness to Pay. Returns a 0-100 score plus a one-line verdict so you know whether to commit a weekend to it or move on. A 60-second triage when you have multiple ideas and need to pick one, not a substitute for the full validation work once you've picked.
Your idea, 5 honest questions
Are you certain there are at least 10,000 buyers in your reachable market?
Can you name a specific person (job title + daily situation) who feels this pain?
Have 3+ buyers told you they already spend money or hours trying to solve this?
Is there 1+ thing about your approach that incumbents can't or won't copy in 6 months?
Have 3+ buyers joined a waitlist or pre-paid, beyond just saying "interesting"?
Your idea score
0 of 5 answered
Answer all 5 questions to see your idea score.
What the number means
| Idea score out of 100 | Reading | What to do about it |
|---|---|---|
| 0 to 39 | Act now | You are working from assumptions rather than evidence, which is a normal place to start and a dangerous place to build from. Spend two weeks talking to buyers. Nothing you code this fortnight will survive those conversations anyway. |
| 40 to 69 | Fragile | There is a real signal here and it is not yet supported. Two weeks of customer development will either lift this above seventy or kill it cleanly, and both outcomes are worth having. |
| 70 to 100 | Healthy | Enough evidence to justify building the smallest useful version. A green light rather than a guarantee: the score measures what you have learned, not what will happen. |
Worked examples
| Situation | Numbers in | Answer out | Verdict |
|---|---|---|---|
| Idea from the shower this morning | 1 maybe, 4 not yets | 10 / 100 | Pre-validation. Everything is still a hypothesis wearing a plan's clothing. |
| Some research, no conversations | 1 yes, 2 maybes, 2 not yets | 40 / 100 | Promising and unsupported. The missing evidence is always willingness to pay. |
| Ten buyer interviews done | 2 yeses, 3 maybes | 70 / 100 | Just over the line. Build the smallest version and keep interviewing. |
| Waitlist with deposits taken | 5 yeses | 100 / 100 | Strong across all five. The risk has moved from the idea to the execution. |
How this is calculated
Each of the 5 questions scores 20 (Yes), 10 (Maybe), or 0 (Not yet). The total is the Idea Score, 0-100.
The 5 dimensions are the ones that show up in every post-mortem of failed pre-PMF startups:
- Market. Is there a real buyer pool?
- Buyer. Can you name a specific persona?
- Pain. Are buyers already paying to solve this?
- Differentiation. Can incumbents copy you?
- Willingness-to-pay. Has anyone committed to pay?
Source mapping: CB Insights "Top Reasons Startups Fail" survey of 110 failed-startup founders. 42% died from no market need (Market + Pain), 17% from being out-competed (Differentiation), 14% from no demand (Buyer + WTP). The 5 dimensions cover the failure modes; the score predicts which dimensions you've validated.
What this doesn't tell you
- Whether you can actually build it. A 100-score idea with no team to execute is still zero. The score validates the idea, not the team.
- Whether the timing is right. Some ideas score 80 today and 30 in 2 years (market shift); others score 30 today and 80 in 2 years. Score + timing both matter.
- The strength of evidence behind each Yes. "Yes" with a paid waitlist of 50 buyers is different from "Yes" with 3 verbal "interesting"s. The calc treats them the same. Use the per-dimension playbook to know which.
How do you know if a startup idea is good?
You cannot know, but you can measure how much of your belief rests on evidence rather than enthusiasm. That is what the five dimensions test: whether the market is real, whether you can name the buyer, whether the pain already costs them money, whether you have any advantage that lasts six months, and whether anyone has paid or queued. An idea nobody has tested is not a bad idea, it is an untested one, and the two are easy to confuse when the idea is yours.
What makes an idea score low?
Almost always the fifth dimension: willingness to pay. Founders can usually name a market, describe a buyer and argue for a difference, because those can be reasoned about at a desk. Evidence that someone will actually part with money requires leaving the desk. This is also why low scores cluster at exactly the moment people feel most confident, which is the fortnight before they start building.
Is a score of 60 good enough to start building?
It is good enough to start testing, which is not the same thing. Sixty means most dimensions are half-supported, and the cheapest next step is a fake-door test: a page, a real price, and a payment button that goes nowhere. If forty of your named buyers see it and nobody clicks, you have saved two months for the cost of an afternoon. If they do click, you now have the evidence that turns two of those maybes into yeses.
Put this on your own site
Free to embed, no permission needed, no tracking script smuggled in. The only condition is the credit link that comes with it, which seems a fair trade for the arithmetic.
<iframe src="https://shipfit.ai/calculators/embed/idea-score" width="100%" height="900" style="border:1px solid #e2e8f0;border-radius:16px;" title="Idea score by ShipFit" loading="lazy"></iframe>Use this with
The Mom Test
The Mom Test is Rob Fitzpatrick's framework for customer interviews that generate real signal. Not praise. Three rules, applied step-by-step, with examples.
Idea Validation
Most founders confuse idea validation with idea-receiving-encouragement. The two have nothing in common. Here's what real validation looks like, and the four methods that actually produce it.
How do I know if my business idea is good?
Run six honest tests. (1) Is there a defined buyer with budget? (2) Is the pain frequent and intense, not annoying? (3) Will buyers commit money or time? (4) Do you have a defensible angle (one of the 7 Powers)? (5) Does the unit economics math work at a price they'd pay? (6) Is the timing present-tense, not 'someday'? Pass four of six and keep going. Fail four and walk away. The rest is theatre.
Pricing strategy calculator
Van Westendorp in 4 numbers. Skip the survey-platform fees.
CAC (Customer Acquisition Cost)
What it costs you, on average, to convert one new customer. Includes ad spend, sales salaries, and the tools that make those people effective. The most-underestimated number in early-stage finance.
Frequently asked questions
What's the difference between this and an MVP test?
Why these 5 dimensions specifically?
What's the difference between Yes, Maybe, and Not yet?
Why is the per-dimension breakdown gated?
A score is a diagnosis, not a plan.
ShipFit takes the same five dimensions and turns them into nine decisions with the evidence attached, ending in a spec you can hand to Cursor or Claude Code.