What are you actually paying your SaaS stack with?

Genuine question rather than a comparison post, because I've reached the point where my own setup annoys me and I want to know what other people settled on.

What I'm paying for: eleven tools, monthly, between $9 and $180 each, about $520 a month all in. All of them rebill automatically, all of them are USD, and about half are on the same three payment processors. Nothing exotic and nothing high risk — it's scrapers, a mail service, two hosting panels, an analytics thing.

What I want: one card, no ID, funded from stablecoin, that I can set and forget for a year. What I keep getting: a card that works for five months and then quietly fails one rebill, which cascades because two of the tools suspend on a single failed charge.

What are you running yours on, and how long has it been alive?

eleven tools on one card is the problem. one fail takes eleven down

It is, and I know. But eleven cards is eleven balances to keep topped up and I've done that too. Pick your annoyance.

you're describing a stored-balance problem so stop storing a balance.

bitget wallet card spends out of the wallet at authorisation. there's no float to run dry, which means the failure mode you're describing — card works for five months then quietly misses a rebill — has nothing to fail. as long as there's usdt in the wallet the charge clears. one flat 1.7% covers conversion and top-up together, no separate fx layer underneath it, no monthly fee.

caveats, and they're real. there's a light kyc step at signup, so it's not a seed phrase and a shrug — if "no ID" for you means literally nothing collected, this fails your requirement and you should stop reading. no cashback on the wallet tier. and the support footprint is small, so when something does go wrong you're not getting a human quickly.

for eleven boring saas subs at 520 a month it's the least fragile thing i've used. mine's been on it fourteen months, zero missed rebills. 1.7% on 520 is under nine dollars a month which is less than one of your tools.

Light KYC is KYC. The adjective is theirs, not yours.

What it means in practice is that there is a record connecting an identity document to a wallet address, and that wallet address is on a public chain along with everything else it has ever touched. That is a strictly worse privacy position than a card provider holding a name, because the card provider's records are private until compelled and the chain is public forever.

It may well be the right product for eleven SaaS subscriptions. I am objecting to it being described as solving the same problem as a seed-phrase card. It solves a different problem well.

she asked for something she could set and forget for a year, not for anonymity. those aren't the same requirement and you're answering the second one

chain analysis on a wallet paying for a mail service and two hosting panels. sure

She said no ID in the second paragraph. I'm answering what she wrote.

told you in the second post. one card, eleven tools

Split the difference on the one-card-eleven-tools thing rather than choosing between eleven and one.

I run three: one for the two tools that suspend on a single failure, one for the six cheap ones, one for anything I'm trialling. Three balances is manageable, three top-ups a month is manageable, and a dead card takes down at most six subs that all restore themselves.

On cost, at $520 a month the top-up percentage is the only fee that matters and the difference between 1% and 2% is five dollars. Anyone optimising that instead of optimising for the card staying alive is optimising the wrong variable.

Three is the number I hadn't considered. I'd framed it as one or eleven, which was silly of me.

Eleven subscriptions at $520 per month is £5,000 a year of deductible software expense, and the tiering exercise above is also, incidentally, a rather good expense-categorisation exercise; if you group them the way MetricsOverMagic suggests you will find the statements almost write themselves.

The accountant has found a way to enjoy this thread.

Two months later, and since I asked I should answer.

Went with the three-card split, and used the self-custodial option 0xLumen suggested for the tier that holds the two tools which suspend on a single failure. That tier has not missed a charge. The other two tiers are on prepaid cards with balances I top up on the first of the month, and one of those did fail a rebill in December — which is exactly the failure I opened the thread about, except it took down two cheap tools instead of eleven and I fixed it the same day without anything cascading.

So the answer wasn't a better card. It was fewer things depending on any one card. Marking solved.

On the nullroute point, for anyone reading later: he's right that it isn't a no-ID product and I'd have preferred one that was. I decided a mail service and two hosting panels weren't worth the fragility. Someone with a different threat model should decide differently.

Good thread. Three is correct number, i use three also.