The float, precisely
Buy inventory on day one of a statement cycle and the charge is not due until the statement closes and the grace period runs, roughly 50 days later. Pay in full and the financing cost was zero. That is the base float, and timing large purchases to the day after statement close maximizes it.
A 0% intro purchase APR stretches the same mechanic across 12 to 18 months. Inventory bought in month one can be sold, restocked, and sold again several times before the first dollar of interest is owed. For a business with reliable sell-through, this is genuinely cheaper than any line of credit on the market, because it is free.
The discipline that makes it work is one rule: sell through before the promo ends. The 0% period is a runway with a hard end date, and the balance at the end reprices to 18% to 29%. Every purchase decision should carry its own answer to the question of which week this inventory converts back to cash.
Paying vendors that do not take cards
The float only covers what you can put on a card, so the rails that turn card credit into vendor payments matter. Melio is the current workhorse: it charges about 2.9% to pay a vendor by card while the vendor receives a bank transfer or check. That fee is the toll for extending float to your whole payables ledger.
Plastiq, the previous workhorse, is the cautionary tale. It filed for bankruptcy in 2023 and businesses that had built their payables process on it scrambled. Treat any bill-pay intermediary as replaceable infrastructure and keep a second rail configured.
Rent has its own rail: the Bilt card pays rent with no transaction fee and earns points doing it, which matters for operators whose largest monthly fixed cost is a lease.
The only equation that matters
Card-funding a payment makes sense when fee < rewards + float value + 0% spread. Work it in basis points. A 2.9% Melio fee against 2% rewards, six weeks of float worth maybe 0.5% at current yields, and no 0% promo is a losing trade. The same fee against an 18-month 0% runway that replaces a 12% line of credit is a clearly winning one.
Most failed float operations never wrote this equation down. They paid 2.9% reflexively for points worth 1%, or they valued the 0% spread against money they would not otherwise have borrowed. If the line item does not clear the inequality, pay it from the bank account like a civilian.
- ▸Fee: the bill-pay or liquidation cost, typically 2.9%+
- ▸Rewards: realistic cents-per-point value, not aspirational
- ▸Float value: days of float times your actual yield on cash
- ▸0% spread: the rate on the financing this replaces, counted only if you would have borrowed anyway
What breaks it
Unsold inventory is the primary failure. The card balance is due regardless of whether the goods moved, and a slow season that pushes a balance past the promo end converts free financing into the most expensive debt a business can carry.
Platform payout holds are the second. Amazon, Shopify, and payment processors can hold funds for weeks on a review trigger or a suspension, and the float math assumed those payouts would arrive before the statement did. A reserve covering at least one full cycle of card payments is not optional.
The third is quieter: utilization. Capital One business cards report to personal bureaus, and a fat inventory balance on any personally-reporting card spikes your utilization and drags your personal score, right when you might need it for the next round of approvals. Keep float balances on cards that do not report routine activity personally, and know which of yours do.
