Most SaaS ideas die for the same boring reason: nobody checked whether anyone would pay for the thing before spending three months building it. Validation isn't a formality you rush through to get to the fun part. It's the fun part — it's the only step where you're trading a few days for the right to skip months of wasted engineering.
Validation is not asking "would you use this?"
Every founder eventually asks a friend, a coworker, or a stranger on Twitter, "would you use this?" Almost everyone says yes. Politeness is free, and hypothetical enthusiasm costs nothing to give. The answer tells you nothing about whether that person would ever open their wallet, switch tools, or change a single habit for your product.
Real validation asks a different question: has this person already demonstrated, with their own time or money, that this problem is expensive enough to solve? You're not looking for opinions about the future. You're looking for evidence from the past — a receipt, a workaround, a canceled subscription, a five-paragraph rant. Behavior beats opinion every time, and behavior already happened, which means you don't need a survey to find it. You need to go read where people are already complaining, comparing, and building duct-tape solutions.
Start from demand you can already see
The founders who validate fastest don't start with an idea and go looking for confirmation. They start with a visible pile of frustration and work backward to the idea. Reddit threads asking "is there a tool that does X," Stack Exchange's softwarerecs tag, 1★ App Store reviews of tools people already pay for, GitHub issues with dozens of thumbs-up on a feature request — these are public, timestamped, and impossible to fake. Someone spent real effort writing them, which is itself a small proof of pain.
This is the inversion that matters: instead of validating an idea you invented, you're mining ideas out of demand that already exists. The problem was never "will people want this" — it's "which of the hundreds of problems people are already describing is worth building for."
The seven signals worth money
Not all complaints carry equal weight. Below is roughly the ladder of buyer intent, from strongest signal to weakest. Weight what you find accordingly — a handful of "take my money" comments outweighs a hundred generic gripes.
- Willing to pay. Explicit language — "would pay for," "take my money," "shut up and take my wallet." This is as close to a purchase intent signal as text gets. Rare, and the most valuable thing you can find.
- Switching. "Alternative to [tool]," "canceling my subscription because," "too expensive for what it does." These people already pay for something in your category — the hardest part of a sale, proving the category is worth money, is already done.
- Solution-seeking. "Is there a tool that…," "what do you use for…" Active, present-tense searches for a solution. High commercial intent, but no proof yet that they'd pay a specific price.
- Workaround. "I built my own spreadsheet for this," "cobbled together Zapier and three apps." Real time investment to solve the problem badly — a strong tell that a well-built version would be adopted.
- Repeated pain. "Still doing this manually," "so tedious every week," "someone needs to build this." Genuine frustration, but unconfirmed willingness to pay — treat it as a lead, not a conclusion.
- Volume and recency. One post is an anecdote. Ten similar posts across different communities in the last month is a pattern. Volume across sources is itself a signal, independent of what any single post says.
- Source quality. A complaint on a professional subreddit (accountants, MSPs, agencies) full of paying-business owners is worth more than the same complaint from a hobbyist forum. Who is saying it matters as much as what they're saying.
Score before you build
Once you've collected signals, don't just eyeball them — score them. A simple mental model: multiply engagement × intent × source weight. Engagement is upvotes, replies, and reactions — proof other people share the pain. Intent is where the post sits on the ladder above. Source weight accounts for who's talking — a thread of small-business owners complaining is worth more than the same volume from students or hobbyists.
This is the same principle BuyerTell runs on: signals aren't treated equally just because they mention a keyword. "Would pay for" is weighted roughly 1.8× a generic pain post, because it's telling you something fundamentally different about the person behind it. If you're doing this manually, keep a spreadsheet and be honest about the weighting — it's tempting to inflate the score of an idea you're already excited about.
The market has already told you what to build. Most founders just never read the signal — they invent a solution first and go looking for validation second.
Cheap experiments that de-risk in a week
Once a problem scores well on paper, spend a week — not a quarter — turning that score into real evidence. None of these require code.
- Landing page + waitlist. Write the one-sentence pitch, put up a page with a single CTA, and drive a small amount of traffic to it — the same communities where you found the signal. A 3-5% email conversion from a cold, relevant audience is a genuinely good sign. Anything under 1% is a warning, not a death sentence, but it means you should re-check whether you're solving the actual pain or an adjacent one.
- Concierge MVP. Do the job manually for 3-5 real people before writing software. If the workflow you'd automate isn't valuable enough for someone to let you do it by hand for a fee, it isn't valuable enough to build a product around.
- Fake door. Add a "Coming soon — join the beta" button inside an existing tool, community, or newsletter, and measure the click-through. This is the fastest way to test demand for a feature without building it — but keep the audience honest and relevant, or the number means nothing.
- DM the complainers. Go back to the actual posts you found and message the people who wrote them. Ask what they're currently paying (if anything), what they've already tried, and whether they'd hop on a 15-minute call. A handful of real conversations will tell you more than a hundred survey responses, because you're talking to people who've already proven they care enough to write publicly about the problem.
- Pre-sell before you build. If you can get even one or two people to pay upfront — a deposit, an annual plan, a "founding customer" rate — for something that doesn't exist yet, you've validated harder than almost any other signal available to a pre-product founder.
This entire process — finding the signal, weighting it by intent, and turning it into a scored, evidence-backed brief — is what BuyerTell runs every morning across six sources. It reads the same threads you'd read manually, scores them the same way described above, and hands you three ranked ideas with the actual quotes attached, so you can skip straight to the week-one experiments. See how the phrases themselves are weighted in buyer-intent signals: the phrases that predict willingness to pay.
The takeaway
Validation isn't a gate you pass through once. It's a habit of trusting evidence over enthusiasm — yours and everyone else's. Find the pain that already exists, weight it honestly, and spend a week proving it with money or behavior before you spend a quarter building around it. The ideas that survive that process are the ones actually worth your time.