AMA — four years running no-KYC cards at volume, ask what you like

Four years in - what's the one thing you believed at the start that turned out to be completely wrong? Not a mistake, a belief.

Had a similar reckoning at about the three year mark myself and it was uncomfortable.

That the providers were the risk.

I spent two years building elaborate defences against a provider vanishing — spread balances, multiple accounts, the monitoring, all of it. That work was correct and I would do it again, and across four years it has cost me $3,180.

In the same period, declines and dead cards on live campaigns have cost me somewhere north of $19,000 in wasted spend, missed windows and campaigns that stopped mid-flight. I have never built anything to defend against that beyond a calendar reminder, because it never felt like a risk, it felt like friction.

The thing I got wrong is that I was measuring the dramatic failure and ignoring the boring one, and the boring one is six times more expensive. That is embarrassing for someone whose entire posting personality is about measuring things.

Late to this. At your volume, has any provider ever asked you a question they didn't have to ask?

Twice. Both times after a single transaction well above my normal pattern rather than because of cumulative volume, which I think is the actual answer to what you're asking.

Neither question came with a hold. Both were answerable with one sentence. Neither provider followed up.

Winding this up after four months since the questions have stopped.

Two I never answered and should note rather than leave hanging. vlad_ptk asked whether I have ever recovered a balance from a living provider who locked it — I have, once, $610 after nineteen days and three messages, and I did not answer at the time because I could not remember the details well enough to be useful and then forgot. quietaccountant asked about bookkeeping and the honest answer is that mine is inadequate and I did not want to write that out in public, which is a bad reason not to answer a good question.

What I'd take away from the whole thread, if anyone reads it end to end: almost every number I posted has a small sample behind it, and the two places where I was most confident going in — SaaS decline rates and whether issuance volume matters — are the two where the data disagreed with me. Treat everything above as one person's records rather than the shape of the market.

Ask me again in 2027 and half of it will be wrong.

half of it was already wrong when you posted it. still the best thread this year