Every one of these is a job a founder actually has to do this week — validate the idea, price the thing, test demand, decide whether to kill it. Tick the steps off as you go. Each step carries the worked example for that exact situation, with real numbers instead of advice — and the free tool that does it, running inline. Going deeper: the SOP library and tool comparisons.
A 7-day test that proves people want your newsletter before you write issue one — no list, no Twitter following, no ad budget.
6 steps · $12The evidence order that kills bad SaaS ideas in days instead of quarters: demand first, willingness-to-pay second, build last.
6 steps · $0Every good .com is gone and .ai renewals cost more than your hosting. The naming and checking loop that ends with a domain you can actually buy today.
6 steps · $70Prove a service sells on evenings alone: the offer, the 30 outreach messages, and the cash threshold that makes quitting a decision, not a leap.
6 steps · $0App stores are where unvalidated ideas go to die quietly. Prove demand with a clickable prototype and 100 real users before you pay a developer a cent.
6 steps · $50No warm intros, no ex-colleagues in the industry, no design partners. The cold-outreach validation loop that gets 5 real buyer conversations in two weeks.
6 steps · $0Pick a first price you can defend, test it on five buyers, and raise it — the pricing sequence for a product with zero usage data.
6 steps · $0The one-page document that decides in advance what failure looks like — written while you are still objective enough to mean it.
6 steps · $0Two-sided means two validations and a chicken-and-egg problem. The single-city, single-category test that proves liquidity before you build anything.
6 steps · $100Extensions are cheap to build and brutal to distribute. Test the distribution before the code — two days, one landing page, one forum thread.
6 steps · $0Runway maths for founders with no investors: what to count as burn, why your salary belongs in it, and the decision date the number really gives you.
6 steps · $0Where the real competitors hide — reviews, job posts, support forums and spreadsheets — and how to read them for the gap you can actually take.
6 steps · $0Presell a cohort before you record a lesson: the free workshop, the twelve paid seats, and the refund promise that makes it honest.
6 steps · $0No ads, no audience, no funnel — the outreach, proof and referral sequence that takes a productized service from zero to ten paying clients.
6 steps · $0Prove people will buy before you order inventory: the ad test, the pre-order page, and the unit-economics check that kills most product ideas on paper.
6 steps · $450A validation landing page has one job and five parts. Build it in two hours, send real traffic, and read the number that tells you whether to continue.
6 steps · $12Bottom-up market sizing you can defend: count real buyers, multiply by a real price, and find out in an hour whether the niche can pay for your life.
6 steps · $0Anyone can rebuild a prompt in a weekend. The validation loop that finds out whether you have a business or a demo — and what the moat has to be.
6 steps · $40Local demand is measurable in a week: search volume you can check for free, a phone number, and ten real quote requests before you buy a single tool.
6 steps · $150Developers ignore marketing and adopt what solves their problem in five minutes. The open-source-first validation loop that finds out if yours does.
6 steps · $0Eight hours a week, a contract you have actually read, and a test designed around scarcity — how to validate without burning out or getting fired.
6 steps · $50Communities die of emptiness, not of bad ideas. Prove twenty people will pay and show up before you open a single channel.
6 steps · $0A two-hour naming process with a hard stop: generate in three shapes, check in batches, screen for collisions, buy one year.
6 steps · $15The app store tells you everything: install counts, review complaints and pricing are all public. Read them properly, then talk to ten merchants.
6 steps · $0Stop comparing them on paper. Score both on the four things that actually predict outcome, then run the cheaper test first.
6 steps · $0Where every dollar goes when you only have a hundred of them — and the free steps that do most of the work anyway.
6 steps · $100The strongest validation signal available — money from strangers — done honestly: what to promise, what to refund, and what the numbers mean.
6 steps · $30Narrow enough to be findable, big enough to feed you. The four filters and the two-week test that prove a niche before you rebrand around it.
6 steps · $0The signals that mean stop, the ones that only look like it, and how to shut down cleanly without wasting what you built.
6 steps · $0Idea generation is a search problem. Four places where paid problems are already documented, and the filter that kills 90% of candidates.
6 steps · $0Free does not mean people will come. The three launch surfaces that reliably deliver a first hundred users, and what to do the day they arrive.
6 steps · $12Five questions, one rule, and the moment you are allowed to describe your idea — how to get information instead of encouragement.
6 steps · $0Job boards are two-sided and cheap to fake. Post twenty real jobs by hand, count the applicants, and find out whether employers will pay.
6 steps · $20Directories are easy to build and hard to monetise. Test whether anyone searches for the list — and whether the listed businesses would ever pay.
6 steps · $15The safest validated idea there is: a spreadsheet with real users. Here is how to find out whether they would pay for the software version.
6 steps · $0Churn kills subscription boxes, not launch demand. Ship three months by hand to fifty subscribers and find out what month four looks like.
6 steps · $400Raise the price for new customers first, grandfather the old ones, read the close rate. The low-risk sequence for an underpriced product.
6 steps · $0Healthcare, finance and legal have real barriers — and that is the moat. How to find out what compliance actually costs before you build.
6 steps · $500A test that answers it in three months instead of three years: work together on something real before anyone signs anything.
6 steps · $0Templates and guides cost nothing to duplicate and everything to distribute. Sell it before you make it, then find out whether it sells twice.
6 steps · $15Measure demand for a feature or product that does not exist — and keep the trust of everyone who clicks. The honest version of the smoke test.
6 steps · $30Cohort charts need thousands of users. Here is what to measure when you have thirty — and the one number that predicts whether anything else matters.
6 steps · $0"We built it for ourselves and others want it" is a promising start and a misleading one. The test that separates a product from a favour.
6 steps · $0The incumbent is not beaten by features. Find the customer they cannot serve profitably, and check they still cannot serve them tomorrow.
6 steps · $0Free tiers need volume, cost money and hide your worst problem. The three questions that decide whether yours earns its keep.
6 steps · $0One selling window a year means one shot at evidence. How to test in the off-season and survive the eleven months that earn nothing.
6 steps · $100Your champion is not the buyer. How to find the economic buyer, arm your champion, and survive security review without a sales team.
6 steps · $0Most podcasts stop before episode ten. Test the format, the guests and your own stamina with three episodes before committing to twenty.
6 steps · $80You do not need code to find out whether people will pay. The manual, no-code and paid-help routes, and when each one is the right call.
6 steps · $100Praise without purchase is a diagnosis, not a mystery. The four causes, how to tell them apart, and what to change for each.
6 steps · $0These workflows are what kitty.build does automatically: nine specialist boards research your idea live, return a cited GO / PIVOT / NO-GO verdict, and then run the build — while nothing outbound goes anywhere without your yes. First idea is free.
Feed her an idea — freeNo card · Failed tasks are free · Your repo, your domain, your revenue