The CB Insights post-mortem analysis of 110 failed startups found that 42% of them failed because there was no market need. Not poor execution. Not bad timing. Not insufficient funding. No one wanted the product.
Most of those founders spent months - sometimes years - building something before they learned this. Validation is the process of learning it in days instead.
Here are the five questions every founder should be able to answer before writing a line of production code.
Why Skipping Validation Is the Most Expensive Mistake in Startups
Building a product takes 3-6 months minimum. Hiring a team, renting infrastructure, and running paid acquisition adds up to tens of thousands of dollars before you know whether the market cares. Validation compresses that feedback loop into days.
The goal isn't to eliminate all uncertainty - it's to eliminate the most costly uncertainty cheaply. You can't know for certain if something will succeed until it's live. But you can know for certain whether people have the problem you think they have, and whether they'd pay to have it solved.
Question 1: Is This a Real Problem, or Just a Frustration?
There's a difference between a mild annoyance and a problem worth paying to solve. A real problem has three characteristics: it's frequent (happens regularly, not once a year), it's painful (creates meaningful lost time, money, or stress), and people are already trying to solve it somehow - even if that "solution" is a broken spreadsheet or a manual workaround.
How to validate this
Search Reddit, Hacker News, and industry forums for your target audience discussing the problem. Count how many unique people describe the frustration. A real problem appears in dozens of independent threads. A mild annoyance shows up once or twice.
Look for DIY solutions - people who built their own workaround in Airtable, Notion, Zapier, or a custom script. Every DIY workaround represents someone who wanted a solution so badly they built their own instead of waiting for you.
Question 2: Are People Actively Searching for a Solution?
A problem people actively search for is a problem with existing buying intent. That buying intent is what makes SEO and paid acquisition viable - people are already looking for what you're building.
How to validate this
Use Google Keyword Planner, Ahrefs free tools, or even Google autocomplete to check search volume for terms like "tool for [your problem]", "best app to [solve your problem]", or "alternative to [existing solution that doesn't quite work]".
For micro-SaaS, 500-2,000 monthly searches for a core term is typically enough to build a profitable business. You don't need mass-market scale - you need a concentrated group of people with a specific, recurring need.
The App Store signal
If there are already apps in your category on the App Store with 1,000+ reviews, that's a market validation signal. People don't write reviews for products they don't use. The question is whether the existing apps serve the market well - the 1-2 star reviews tell you where they fall short.
Question 3: Would Anyone Pay for This?
Free users and paying customers are different species. Something can be popular as a free tool but completely uncommercial. The test is whether the problem is painful enough that people will open their wallets to solve it.
How to validate willingness to pay
The gold standard is a pre-sale: put up a landing page, drive traffic to it, and see if people enter their credit card before the product exists. But you can also gather signals earlier:
- Community responses: Post in relevant subreddits or Slack groups: "I'm building [X]. Would you pay $Y/month for it?" The ratio of genuine interest to skepticism tells you a lot.
- Existing paid alternatives: If people are already paying $50-200/month for a tool that only partially solves the problem, that's direct evidence of willingness to pay.
- Job postings: Companies hiring people to manually do what your software would automate are paying for the solution already - just in labor costs rather than software.
What price to charge
For B2B micro-SaaS, the sweet spot in 2025 is $29-99/month per seat for SMB tools. Anything that saves 2+ hours of work per week can justify $50/month. Anything that prevents a costly error or generates leads can justify $99-200/month.
Question 4: How Big Is the Market?
You don't need a billion-dollar market. A micro-SaaS targeting 5,000 businesses at $49/month has a $2.9M ARR ceiling - more than enough for a profitable solo business. The question is whether your specific niche has enough people with the problem to reach your revenue goals.
The bottom-up market size calculation
- Estimate the total number of potential customers (e.g., "freelance designers who invoice clients")
- Multiply by your realistic market share (1-5% for a new entrant)
- Multiply by your planned monthly price
- Multiply by 12 for ARR
If that number is below $100K ARR at 5% market share, the market may be too small. If it's above $1M, you have room to build a real business. This is a rough estimate - refine it with actual search volume data and competitor revenue signals.
Question 5: Can You Build the Simplest Version in 4 Weeks?
This is the most underrated validation question. Speed to first customer determines whether you can test your other assumptions before running out of runway - money or motivation.
The "simplest version" rule: identify the single action your product takes that creates value, and build only that. Everything else - user accounts, team features, integrations, mobile apps - is a distraction until you have one customer paying.
How to scope a 4-week MVP
List every feature you want to build. Now draw a line through everything except the core value-creation function. That's your MVP. If you can't deliver value without a feature, it stays. If the value survives without it, cut it.
A working spreadsheet or manual process that delivers the outcome counts as a "product" at this stage. Find one customer and do it manually for them. If they pay, automate it. If they don't, you've saved 4 weeks of building the wrong thing.
Where to Find Answers to All Five Questions
The research to answer these five questions traditionally requires 20-40 hours of manual work: browsing subreddits, reading App Store reviews, analyzing competitor pricing, running keyword research. Most founders skip it - not because they think it's unimportant, but because it's genuinely tedious and time-consuming.
This is exactly the problem that IssueMiner was built to solve. The platform continuously monitors Reddit (9 subreddits), Hacker News, App Store reviews, and Stack Overflow, scores each pain point for business viability, and surfaces the ones with the highest opportunity scores. It pre-answers Questions 1, 2, and 3 for hundreds of different problem categories - before you've spent a single hour on research.
The Hobby tier is permanently free. No credit card, no trial expiration. Start finding validated ideas in the next 10 minutes.
The Validation Framework in One Page
- Q1 - Real problem? Find 50+ people describing it independently on Reddit/HN
- Q2 - Active search demand? 500+ monthly searches for related terms
- Q3 - Willingness to pay? People paying for a worse alternative, or building DIY workarounds
- Q4 - Viable market size? $100K+ ARR reachable at 5% market share
- Q5 - Buildable in 4 weeks? Core value delivery with zero extras
Pass all five? Build. Fail two or more? Keep searching. The market that passes all five tests will be dramatically easier to sell than one you talked yourself into.
Frequently Asked Questions
How long should validation take?
With the right tools, the research phase (Questions 1-3) should take no more than a week. Questions 4-5 take another 1-2 days. If validation is taking longer than two weeks, you're probably over-analyzing - or the idea isn't passing the tests and you're trying to rationalize it.
Is it okay to start building while still validating?
Yes - but only build throwaway code. Wireframes, landing pages, and proof-of-concept scripts are fine. Don't write production code, set up a database, or build auth until you have at least one person who said they'd pay.
What if my idea is too niche to validate?
Niche is good. The mistake is going so niche that the total addressable market can't support your revenue goals. A tool for "independent bookkeepers in the EU" might be niche enough to dominate with minimal competition - do the math first.
How does IssueMiner help with validation?
IssueMiner surfaces pain points that have already passed a preliminary validation: they appear across multiple platforms, meet a minimum post-frequency threshold, and are scored by AI for commercial viability. It's not a replacement for talking to customers - but it's a powerful starting point that filters thousands of conversations down to the ones worth investigating.
Next Steps
Pick one of your existing ideas and apply the five-question framework to it today. If it passes, great - move to landing page validation. If it doesn't, useIssueMiner's dashboard to find one that does.
The founders who ship profitable products aren't smarter than the ones who don't. They're just more rigorous about finding problems that people actually have, and more disciplined about building only what's needed to solve them.