01Why asking the question doesn’t work
Ask someone what they’d pay for a product that doesn’t exist yet and you get a hypothetical answer to a hypothetical question — usually generous, because saying a big number costs them nothing. The same person who says “oh, I’d easily pay $50 a month” closes the tab at the checkout page.
Pricing answers only mean something when they’re attached to a real decision someone already made. The good news: people make and describe those decisions constantly, in public, without being asked — you just have to read them where they happen instead of recruiting them into a survey.
02Where willingness to pay actually leaks out
The strongest signal isn’t “I would pay” — it’s “I am paying” or “I paid”. Someone grumbling that a tool costs $40 a month and is barely worth it has told you two things: there’s a market at $40, and there’s an opening below it.
The next strongest is effort. People who’ve built a spreadsheet, stitched together three free tools, or pay a VA to do a thing by hand are revealing demand more honestly than any survey respondent — they’ve spent real time or money rather than nodding at a hypothetical.
03The signals that count — strongest first
Not every mention of money is equal. Ranked by how much they actually predict spend:
- Already paying — “I’ve been on their $29 plan for a year” names a live price the market accepts.
- Paying for a workaround — a VA, a freelancer, or a second tool to patch a gap means budget already exists.
- Switching over price — “I left X when they raised prices” marks the ceiling and the trigger.
- Built it themselves — a DIY spreadsheet or script is unmet demand with sweat already invested.
- Asking where to pay — “is there a tool that just does this?” is a buyer with a wallet out.
- Pure complaint, no spend — real pain, but the weakest willingness-to-pay signal; weight it accordingly.
04Turning scattered signals into a tier
Reading a hundred of these by hand gives you a vibe, not a number. To decide anything you need every mention reduced to the same scale — which is exactly what the pipeline’s willingness-to-pay field does.
Each thread is bucketed into one of four tiers — high, medium, low, or none — based on whether people describe paying for a fix, asking for one, or merely grumbling about the problem. Constraining it to a fixed four-value vocabulary is what lets a report say “31% of threads show medium-or-higher willingness to pay” instead of handing you 300 one-off opinions to re-read.
05Price down from the evidence, not up from your costs
Once willingness to pay is a sortable field, pricing stops being a guess. You can see the prices people already name, the tools they already left, and the gap between “high pain” and “anyone actually paying” — which is usually where a new product lives.
It won’t hand you an exact number; nothing short of a live checkout will. But it replaces “what feels right” with “here is what this audience already pays and complains about”, which is a far better place to run your first pricing experiment from.
See how the scoring works
Willingness to pay is one field in a fixed schema every thread is scored against — here’s the whole method.
How we score pain and willingness to pay →06A worked example — triangulating one number
Collect what they already pay for alternatives
You notice freelancers naming a bookkeeping app at one monthly figure and a stand-alone reminder add-on at a smaller one. That tells you a budget band already exists for “getting paid faster” and roughly where it sits.
Collect what they’ve quit over
A handful describe leaving a pricier suite the moment its renewal jumped, and others abandoned a free workaround once it broke. The quit points mark a ceiling and the triggers that push people past it.
Collect how they value the cost of the problem
People describe chasing late invoices for hours each month, or eating the occasional unpaid one. That’s real time and money the problem already costs them — the upper bound of what a fix is worth.
Form a hypothesis from the overlap
Anchored under the suite they quit, above the bare add-on, and well below the hours the problem burns, you land on a candidate tier — not a verdict.
Test it at a real checkout
Put the number in front of buyers and watch who actually pays. The signals chose where to start; only the checkout settles it.
Frequently asked questions
Read the prices they already act on rather than the ones they predict. Find what they currently pay for alternatives, what they’ve cancelled or switched away from, and what the problem costs them in time or lost money. Those three signals bracket a starting price — a hypothesis you confirm only by putting it in front of a real checkout.
Weight what people do over what they say. A live subscription they grumble about, a freelancer or second tool they pay to patch a gap, a switch made over a price hike — each is a revealed decision worth far more than a survey answer. Score each mention on one scale so dozens of scattered comments become a number you can sort and compare.
Watch where buyers name real figures unprompted — renewal complaints, plan comparisons, “I’m on their cheaper tier” asides, and posts about leaving over a price increase. Those mentions give you actual accepted prices and the ceilings that trigger churn, which is far more reliable than a competitor’s pricing page that may hide discounts and negotiated deals.
Price down from evidence, not up from your costs. Use research to map what the audience already pays, the gap between high pain and anyone actually paying, and the workarounds they tolerate. That sets a defensible first tier to test. Research narrows the range and kills bad guesses; a real checkout, not the research, sets the final number.