Claude ate my product's main feature. I kept the part it can't eat.
june 2025 i launched an ai contracts tool. #6 on product hunt, 141 upvotes, contracts generated from a description. it was the whole product.
six months later claude wrote a better contract than my templates did, and every freelancer already had it open. my core feature became a prompt. i didn't lose to a competitor, i lost to the model.
here's the thing i didn't expect: the parts of the product that still worked were the boring ones. the hours tracked on a mac. the invoice record. the signature. the bank deposit that matched itself to the invoice on tuesday. none of that is generatable.
it's state, and state has to live somewhere.
so i inverted the product.
claude is the interface now. one mcp url, 30 tools, free.
accordio is where the business lives: it measures the hours, holds the invoices and contracts, collects the signature and the payment, and keeps working on whatsapp when claude is closed.
one rule i argued with myself about for weeks: claude drafts, but NEVER sends.
every invoice or contract opens in accordio and you press send. it costs one click every time. it also means the model can't email a client something wrong with your name on it.
honest numbers: solo, self-funded, a handful of paying accounts, 31 signups in 14 days from a $100/day chatgpt ads test. relaunched on product hunt today with the new positioning: https://www.producthunt.com/products/accordio
question for people building on top of models: what part of your product is state rather than generation? that's the part i'd bet the company on now.
The "draft but never sends" rule solves a measurement problem you might not have named yet: it makes the model's failure modes visible.
When contracts are generated and sent, you measure "did the customer complain" - but that's only the biggest failures. When every draft lands in Accordio first, you get something better: "what percentage of AI drafts needed human edits before sending."
That's not a bug tracker. That's a signal measure. It tells you whether the model is actually getting better, and which customer types trust the model more. Hours logged, invoices, signatures - those are state. But the ratio of "model draft to human-corrected output" is the measure that tells you if this business lives or dies.
The state thesis only matters if customers behave differently after the pivot. Of the new accounts, are the paying users actually relying on invoicing, signatures, payments and tracked hours, or are contracts still the reason they came in?
I'd keep the "drafts but never sends" rule even if the model eats the next feature. Writing the draft is getting easy. The expensive mistake is still something going out with your name on it before a person checks it. Hours logged, invoices, anything that has to match a bank deposit later — once it's recorded, you can't undo it with another prompt.
The website only few hundred people saw: https://accordio.ai