SaaS Idea Validation Checklist: 15 Checks Before You Write Code

Most SaaS postmortems trace back to the same root cause: the founder validated the idea in their head instead of in the market. This checklist is the antidote — 15 checks, in the order to run them, each with a concrete pass/fail standard.

It pairs with our guide on how to validate a SaaS idea with Reddit, which covers the community research technique several of these checks depend on.

Layer 1: Does the Problem Actually Exist?

1. You can state the problem in one sentence a stranger would recognize

Write it as: "[Specific person] struggles to [specific task] because [specific obstacle]." If you need a paragraph, the idea is a solution looking for a problem.

Pass: A person matching the profile says "yes, that's a real thing" without prompting.

2. People complain about it in public, repeatedly, without being asked

Search the Reddit communities, forums, and Slack groups where your target user talks shop. You are looking for the problem described in their words, across multiple threads, over months — not one viral complaint.

PainPointMap automates this check: scan the subreddits where your target user is active and it returns recurring pain points ranked by frequency, with links to source threads.

Pass: 10+ independent complaints across 6+ months. Fail: You searched three communities and found near-silence.

3. The problem recurs — it is not a one-time annoyance

Subscription software needs a recurring problem. A pain someone hits once a year does not sustain a monthly invoice.

Pass: The problem appears weekly or monthly in your target user's workflow.

4. Users have built workarounds

Spreadsheets, Zapier chains, hired VAs, duct-taped scripts. Workarounds are the strongest problem evidence there is — people invested real effort because no product solved it.

Pass: You find at least three descriptions of manual workarounds in community threads.

Layer 2: Is the Market Real?

5. Buyers already pay for something adjacent

The easiest customer to win pays for an inferior solution today. A market with zero paid products usually means zero willingness to pay, not an untouched goldmine.

Pass: At least one existing paid tool, service, or consultant serves this problem.

6. You can name and count the buyers

"SMBs" is not a market. "Independent insurance agencies with 2-20 staff in the US (~35,000 of them)" is. If you cannot estimate the count, you cannot judge whether the niche supports a business.

Pass: A named segment with a defensible size estimate.

7. The buyer and the user are reachable

List the specific channels: which subreddits, which newsletters, which conferences, which job titles on LinkedIn. If no channel exists where these people congregate, customer acquisition will be brutally expensive regardless of product quality.

Pass: Three named channels where your target buyer demonstrably pays attention.

8. The problem costs the buyer real money or time

Quantify it: hours per week lost, revenue leaked, penalty risk. Your price ceiling is a fraction of this number.

Pass: The problem plausibly costs 5-10x your intended annual price.

Layer 3: Can You Beat the Alternatives?

9. You have read the complaints about every existing competitor

Sort competitor reviews by 1-3 stars. Search "[competitor] alternative" in the target communities. The recurring complaints are your product spec and positioning, pre-written by the market.

Pass: A consistent complaint pattern that your approach addresses directly. See how to research competitors on Reddit for the full method.

10. The dominant alternative is beatable — including "do nothing"

Your real competitor is often a spreadsheet or the status quo. Ask: why would someone switch from the thing that is currently free and familiar?

Pass: A one-sentence answer to "why switch" that references the cost of the current alternative.

11. Incumbents cannot copy the fix trivially

If your entire edge is one feature the incumbent could ship in a sprint, the window is short. Durable angles: a narrower segment the incumbent will not prioritize, a business model they cannot match, or a workflow position they do not occupy.

Pass: Your differentiation survives the question "what happens when the market leader notices?"

Layer 4: Will Anyone Actually Pay?

12. Prospects rank the problem in their top three

In conversations, do not pitch — ask what their hardest problems are. If your problem never makes their unprompted top three, you are building a vitamin, not a painkiller.

Pass: 3 of 10 prospects raise the problem area unprompted.

13. A landing page converts cold traffic

A one-page description of the offer with an email signup, put in front of the target audience. Not proof of purchase intent, but a cheap filter.

Pass: 10%+ visitor-to-signup from targeted traffic.

14. Someone commits before the product exists

The gold standard: a paid pre-order, a signed pilot agreement, or a deposit. Even at a discount, even refundable — a payment converts opinion into evidence.

Pass: One real commitment. Strong pass: three.

15. You would still want this business in year three

Validation is not only market-side. The niche, the customer type, and the problem will be your daily life for years. Support tickets from these users, sales calls with these buyers.

Pass: An honest yes.

Scoring

• 13-15 passes: Build. Start with the smallest version that delivers the core promise.
• 9-12 passes: Investigate the failures — most are fixable by narrowing the segment or repositioning.
• Under 9: The market is telling you something. Run the next idea through the list; you just saved yourself a year.

Related Reading

• How to Validate a SaaS Idea Using Reddit — the community research method behind checks 2, 4, and 9
• How to Validate a Business Idea Using Reddit — the same approach for non-SaaS businesses
• Idea Validation Framework — the conceptual model this checklist operationalizes
• What Happens If You Skip Validation — the failure modes this list exists to prevent