Three gates and six weighted tests, each with a specific threshold. Run them in order and stop at the first hard fail.
A SaaS idea validation checklist is only useful if every item is pass or fail. Nine tests below, each with a specific threshold. Run them in order and stop at the first hard fail.
Nothing here requires building anything. That is the point.
The failure mode this checklist prevents is the gradual one: an idea that never clearly fails, never clearly passes, and absorbs eight months.
Test 1 — Ten people describe the problem unprompted. In ten conversations with people who should have it, count how many raise it before you do. Fewer than three is a fail. Recognition when prompted is not demand.
Test 2 — Someone already pays to solve it. A competitor, a freelancer, an agency, or staff hours. Any existing spend passes. Zero spend anywhere is a fail, because you would be creating a budget line rather than redirecting one.
Test 3 — The problem recurs at least monthly. Subscription revenue needs a recurring problem. Annual or one-off problems fail this gate regardless of how painful they are, because customers finish and leave.
Fail any of these three and stop. The remaining six tests cannot compensate, and continuing past a failed gate is the most common way founders spend a year on an idea that was answerable in a fortnight.
| # | Test | Pass threshold |
|---|---|---|
| 4 | You can reach these buyers without permission | A named forum, subreddit, list or event |
| 5 | Someone commits something real | 3 of 10 offer a card, pilot or date |
| 6 | The value shows in one session | First useful output inside 30 minutes |
| 7 | You can build a first version in 6 weeks | Alone, part-time |
| 8 | The buyer decides alone | No procurement, no committee |
| 9 | You understand the domain | Better than most who might build it |
Four or more passes means proceed. Two or three means the idea is weak but salvageable by narrowing the segment. Fewer than two means you are choosing difficulty for no compensating advantage.
Commitment separates real validation from pleasant research, and it is the test founders most often skip because it risks a clear no.
Ask for one of four things: a pre-order, a paid pilot, a diary date to trial it, or an introduction to whoever controls the budget. Each costs the person something, which is what makes the answer informative.
Ten enthusiastic conversations with zero commitments is a negative result rather than a neutral one. If that is your outcome, it is worth reading the kill signals before continuing.
Test 8 sounds like a detail and decides your entire sales motion.
If your buyer needs sign-off from a manager, procurement, legal or IT, your sales cycle stretches from days to months and you cannot run it self-serve. That is a viable business and it is not a solo one — it needs someone available for calls at the buyer's convenience.
Individual operators, freelancers, and small teams where one person owns the budget are the segments a solo founder can serve without a sales function.
Nothing on this checklist requires a product. If your instinct is that you need to build first to test properly, that instinct is what turns a two-week validation into a six-month one.
| Result | Meaning | Do this |
|---|---|---|
| 3 gates + 5–6 weights | Strong. Build it. | Set pricing, then a launch date |
| 3 gates + 4 weights | Proceed with a narrower segment | Pick the segment that passed test 4 |
| 3 gates + 2–3 weights | Weak but salvageable | Re-run with a different buyer type |
| Any gate failed | Stop | Take the learning to the next idea |
If you are running this against two competing ideas rather than one, score both and compare — choosing between two SaaS ideas covers the tie-breakers when the scores come out close.
Pricing before building. The interviews told you what the current alternative costs, which is the input that sets your ceiling — the pricing framework covers which method applies at zero customers.
Then decide what unit you charge for, because it is harder to reverse than the price and it determines whether revenue grows as customers succeed. The value metric framework covers the four tests.
If you have not yet run the conversations this checklist depends on, start with validating with no audience, and see the full validation process for the wider sequence.
Three gates and six weighted tests. The gates: ten people describe the problem unprompted, someone already pays to solve it, and the problem recurs at least monthly. Fail any gate and the remaining tests cannot compensate.
Two to three weeks, almost all of it conversations. Nothing on a proper validation checklist requires building a product — the instinct that you need to build first is what turns a two-week validation into a six-month one.
Commitment. Ask for a pre-order, a paid pilot, a diary date or an introduction to the budget holder. Three commitments out of ten conversations is a strong signal, and ten enthusiastic conversations with zero commitments is a negative result rather than a neutral one.
If your buyer needs sign-off from a manager, procurement or IT, the sales cycle stretches from days to months and cannot run self-serve. That is a viable business but not a solo one, because it needs someone available for calls at the buyer's convenience.
All three gates passed plus four or more of the six weighted tests. Three gates with two or three weights means the idea is weak but salvageable by narrowing the segment. Any failed gate means stop.
Describe your idea and what you have tested. Marcus runs the checklist and tells you which gate you are closest to failing.
Try GhostCoach free →14-day free trial · cancel anytime · 30-day money-back on Lifetime