How to validate a SaaS idea before building it
Almost every failed solo SaaS was validated by asking people whether an idea sounded good. It did. Then nobody bought it.
The short version
- \"That sounds useful\" is not validation. It costs the person nothing to say and predicts nothing.
- Look for an existing workaround. A spreadsheet, a VA, or a tool they resent paying for means the budget already exists.
- Ten conversations beats a hundred survey responses, because you can ask the follow-up question.
- The landing page test takes an afternoon and answers more than a month of thinking.
- Validate the buyer, not the idea. A good idea for nobody in particular is worse than a dull one for a defined group.
What does validating a SaaS idea actually mean?
Finding evidence that a specific group already spends time or money on the problem, before you spend months on the solution. Not evidence that people like the idea; evidence that they are already paying, in hours or in cash, to avoid the thing you would fix.
The distinction matters because liking is free. Ask ten people whether a tool that saves them two hours a week sounds good and ten will say yes, and none of them has committed anything. Ask what they do about it today and you will get either a real answer (a spreadsheet, a contractor, a tool they hate) or an admission that it is not actually a problem.
A workaround is a budget with a bad interface.
Where do you find people who have the problem?
Where they complain in public: two- and three-star reviews of competing tools, community threads, and job postings. All three are written by people describing a problem in their own words, without knowing you are reading.
| Source | What it tells you | Why it is underused |
|---|---|---|
| 2–3 star reviews | Exactly what a product fails to do | People read 5-star reviews instead |
| Community threads | The words your landing page should use | Feels like procrastination; it is research |
| Job postings | A published budget for solving it manually | Almost nobody thinks to look |
| Competitor changelogs | What the market is asking for loudly | Assumed to be marketing noise |
A company hiring a part-time person to reconcile spreadsheets every week has told you, in public, what they will pay to make that stop. That is a stronger signal than any survey.
What questions should you ask?
Questions about the past, not the future. "What did you do the last time this happened?" and "what did that cost you?" produce facts. "Would you use a tool that…" produces politeness, and politeness has bankrupted more founders than competition has.
- Walk me through the last time you dealt with this. Specific and recent beats general and hypothetical.
- What do you use now? The answer names your real competitor, which is usually a spreadsheet.
- What does that cost you, in money or hours? This is the number your price will sit against.
- What have you tried that did not work? Tells you which obvious solution to avoid.
Never pitch during these conversations. The moment you describe your idea, the person switches from reporting their life to evaluating your feelings, and every answer after that is worthless.
Is a landing page enough to validate?
It is the cheapest useful test, if you send real traffic to it and measure a real action. Write the page before the product (the outcome, the price, the one workflow) and send it to twenty people who have publicly described the problem. If nobody asks when it ships, you learned that in a day instead of a quarter.
What counts as a signal, in ascending order of strength: an email address, a reply asking a specific question, a request for a call, and a pre-payment. Only the last two are strong, and only the last one is proof.
What does not count: page views, time on page, and encouraging comments from other founders. Founders are the most encouraging and least representative audience available to you.
How do you know when to stop validating and start building?
When you can name the buyer, describe the problem in their words, and say what they currently pay to avoid it. If those three sentences exist and are specific, more research is procrastination wearing a lab coat.
There is a failure mode on this side too. Validation is comfortable: it feels productive, it involves no risk, and there is always one more conversation to have. Set a number of conversations in advance, ten is usually right, and commit to deciding at the end of them.
Frequently asked questions
How many people should I talk to?
About ten, if they are the right ten. You are looking for a repeated pattern, and patterns show up fast when the group is specific.
Can I validate without talking to anyone?
Partly. Reviews, forums and job posts are people describing the problem without you in the room, which removes the politeness bias entirely.
What if a competitor already exists?
Good. A competitor is proof that people pay for this, which is the hardest thing to establish. Read their bad reviews and build the thing those reviewers are asking for.
Should I pre-sell before building?
If you can, yes: it is the only validation that involves the person losing something. Be explicit about the timeline and refund on request, or you are selling a promise you might not keep.
How long should validation take?
One to two weeks of part-time work. Longer than that and you are usually avoiding the decision rather than informing it.
The system behind this, written down
Everything above is the map. The Income Loop is the work inside it: modules 0–6 from the problem you solve to the offer that pays for it, plus ten traffic paths: the deeper post banks, the content sales systems and the full software build sequence, in one place.
Get The Income LoopNot ready to pay for anything? The Basic Income Loop is free and includes a complete seven-day starter for Threads, Instagram or software, enough to find out which one suits you before spending anything.