The SaaS Idea Validation Checklist: 25 Questions to Answer Before You Build
July 19, 2026 · 8 min read
Idea validation fails in a predictable way. People validate the part that is fun to validate, usually "is the problem real," and skip the parts that actually kill products: whether the pain repeats, whether anyone will pay before it is polished, and whether you can reach these people without a budget.
This checklist covers all six. Every item comes from a mistake or a win in the They Move the Needle interviews, where founders describe how they got their first paying customer and what they would redo.
Work through it before you commit months. Score at the end.
Stage 1: The problem
1. Can you describe the problem as a single moment with a cost?
Not "creators struggle with distribution." Instead: a Medium account earning around $8,000 a month goes under DMCA investigation overnight and years of audience sit inside a platform you do not control. That moment is where Narrareach came from.
2. Do you know what people do about it today?
The current workaround is your real competitor. Usually it is a Google Doc, a spreadsheet, or nothing. DraftKit's users said plainly that without shared writing they would go back to Google Docs.
3. Does the problem recur, or is it a one time fix?
CheckVibe's early users scanned once, fixed everything, and left. Its co-founder built a security tool and accidentally built a one time transaction. He rebuilt around continuous monitoring, because code keeps changing and new issues keep appearing.
4. What happens if nobody solves it?
If the honest answer is "life continues, slightly annoying," you are building a vitamin. That can still work, but only with very low friction and very low price.
5. Have you felt this problem yourself, and have you checked whether that biased you?
PocketHog exists because its founder wanted to know which of her four apps needed attention, and she used it every day. DraftKit's founder felt scheduling pain and built the wrong product around it. Personal pain is a strong start and an unreliable finish.
Stage 2: The customer
6. Can you name five real people who have this problem?
Names, not segments.
7. Do they gather somewhere you can reach for free?
Subreddits, Slack communities, Discord servers, niche forums. PocketHog used the PostHog subreddit. GainFrame used the Hevy subreddit. She Shapes Digital used a Slack community for women in tech.
8. Can you describe the problem in their words rather than yours?
DraftKit's ongoing struggle: people see a calendar and think booking tool. If your one sentence gets misread consistently, the sentence is wrong.
9. Is your customer defined by a behavior rather than a demographic?
"People who ship AI generated apps to production" is a customer. "Developers aged 22 to 35" is not.
10. Is the segment big enough to matter and narrow enough to reach?
PocketHog's founder is candid that the app is only useful if you already use PostHog. That is a real ceiling, and worth knowing before rather than after.
Stage 3: Demand
11. Has anyone asked you for this before you offered it?
A pull signal beats a polished pitch.
12. Do people say yes to a rough prototype?
She Shapes Digital invited a Slack community to a prototype it said would not be perfect. Eighteen signed up in days.
13. Have you watched at least five people try to use it?
The line from Narrareach: it is uncomfortable watching users use your product, and you learn a lot.
14. Do you know your top three drop-off points?
PocketHog's signup required users to fetch a PostHog API key, and people were dropping off. The founder replaced it with OAuth, and then analytics showed the OAuth flow failing 20% of the time.
15. Are analytics and billing wired up before launch, not after?
CheckVibe's strongest recommendation. You cannot fix conversion or churn problems you cannot see.
Stage 4: Willingness to pay
16. Has anyone paid without being prompted?
DraftKit's first credit purchase came from a free user in the founder's newsletter community who wanted the SMART Draft feature. No launch day sale, no push. That is the highest quality signal there is.
17. Have you tested payment before the product feels finished?
CheckVibe got its first payment before properly launching, off a few TikTok comments with two or three likes each. Narrareach's first payment came a week after launch, when the product was barely functional and the founder considered refunding it.
18. Does your free tier show the problem rather than solve it?
CheckVibe lets anyone scan to see how many issues they have and charges only if there is something to fix. Show the problem first and the paywall makes sense.
19. Do you know why they would pay rather than just why they would use it?
People do not buy polished software, they buy outcomes.
20. Is your pricing model matched to how the value arrives?
DraftKit is running credits while its founder watches how writers actually use them over time, rather than locking in a subscription early.
Stage 5: Distribution
21. Do you have one channel you will still be using in six months?
She Shapes Digital chose LinkedIn even though neither founder enjoys posting there, and their own verdict is that a strategy you dread is a strategy you will not keep up.
22. Can you get in front of these people without paying for it?
Narrareach started SEO from day one, which landed its first users. GainFrame went from Reddit to a blog and TikTok because Reddit is not sustainable forever. One post on generative engine optimization now drives real traffic, and ChatGPT sends 31% of its users.
23. Does the product spread through normal use?
DraftKit spreads through the collaborations it exists to create. Every published joint post exposes it to a new audience.
24. If your personal network vanished, would you still have customers?
DraftKit's founder is honest that her 70 users mostly came from her newsletter, which is great validation and a ceiling.
25. Do you know what your first ten customers will cost you to acquire?
Certado Suite ran for years without KPIs or ROI control, and that is the thing its founder would change first.
Scoring
- 20 to 25 yes: you are validated enough to build the next version properly. Keep instrumenting.
- 13 to 19: the problem is probably real and something else is not. Look at which stage your no answers cluster in, since that is your actual risk.
- Under 13: do not write more code. Do four weeks of customer discovery instead.
The item most people skip
Stage 1 question 3: does the problem recur. It is the difference between a product and a transaction, and it is almost impossible to see from inside the idea. You need to look at what happened to other people who tried the same premise.
Run the pre-mortem before you build.
IdeaGrit runs your idea against red flags and shows six real products that failed with a similar premise, plus a report and an actionable roadmap.
Validate your idea free →