Find out if your idea should work before you build it. A Proof Kit run reads what your future customers complain about in forums and reviews, shows what businesses like yours earn from our own records of 220,000 real businesses, then suggests where to improve and how to reach those customers.
Run your idea · free complaints check
Your questions, answered in counts and quotes, each with its source. Where it found nothing, it says so. The free run shows the complaints and their links. A Validation Kit answers six. A Proof Kit answers all eight.
One word: Strong, Mixed, Not yet, or Not enough evidence. A score out of 100 from up to five checks, each with its count. It can say no.
Real quotes, for and against: the complaints, and what keeps people on what they use now. Every quote links to its thread. Our guesses sit apart, marked.
What the businesses closest to your idea make, in six revenue bands, from our own records of 220,000 real businesses. A band, never a figure we made up.
Can you charge, and how much. Prices by who named them: a rival charges, a page listed, a buyer paid. If no buyer named one, it prints none.
Five steps to tick off, each linked to its proof. Some come ticked when the run already answered them. First: reach your users. Last: check what we couldn’t.
Live threads where someone already asked for a fix, each with a drafted reply you edit and post yourself. It never sends anything: no email, no messages.
How many findings link to a source you can open, and how many it cut or ranked lower. “Found none” is not “never looked.” A thin answer stays thin.
Up to five rewrites of one line you typed, each with its check and quote. None is tested yet. Trying them is a new run; it spends credits.
A Validation Kit run spends 10 credits. A Proof Kit run, the full report, spends 20. An Evidence Pass run spends 3. Your first Evidence Pass is on us. Buy credits
Theirs is a number a language model made up. Ours is a score out of 100, built from five checks: complaints we traced to a page, communities, prices a buyer named, rivals, and businesses like yours that make money. Four are plain counts. The fifth also weighs how strongly people complained, read from the quotes. Every point sits next to what it came from. A Proof Kit runs all five. A Validation Kit runs the complaints, prices and rivals, and scales those three to 100. The report ends with changes we suggest to your brief, each aimed at one check. None is tested yet. You can argue with the call and still keep the proof.
Public talk: Reddit, small forums, reviews and comment threads. Every quote links to the thread it came from, so none of it rests on trusting us.
Then the report says so. A check that finds nothing scores 0, and the report says how to earn those points. What it could not cite is kept apart from what it could. Finding nothing is a real result, and the cheapest one you will ever get.
No. It runs on our side, so you can close the tab and come back to it.
No. Runs happen on our side. Write the idea, pick a depth, go.
No. It cannot post anywhere, and it cannot send email or messages. It finds public threads where someone asked for a fix and drafts a reply. You read it, edit it, and post it yourself, from your own account.
No, and there is nothing to give you. The business records are anonymised: no names, no websites, no contact details of any kind. Nothing in them can point to one company. They exist to work out what businesses like yours earn, and a range is what comes back.
No. Runs do not build one right now. You get the research, the source behind every claim, and the threads where these buyers already talk.
Both, from Settings. The export is everything we hold on your account. The deletion removes your runs and reports for good. It does not hide them.