This is a real end-to-end purchase of this product, run on Base Sepolia with test tokens before mainnet launch. The figures are copied from the run, not illustrative.
A throwaway buyer wallet was funded with 5 test USDC and no ETH. The purchase was driven by the reference x402 client with an explicit 5 USDC spend authorization.
[1] GET /offer 200 offer advertises 5 USDC
[2] POST create order 200 order created
[3] GET /product 402 challenge returned, no content leaked
[4] sign authorization nonce 0x953da366...
[5] GET /product + signature 200 18 files delivered, checksum matched
[6] verify on chain Transfer + AuthorizationUsed matched the order
[7] repeat download 410 purchase already used, no second charge
[8] interrupted settlement 202 refused to re-chargeSettlement transaction: 0xb8762176767d188bdbe40e3a3fdca6d46a78e35d8013d671155f32b0794102b6
Exactly 5.000000 test USDC moved from the buyer to the receiving address. The buyer's balance ended at zero, confirming a single charge. The delivered body hashed to the value in the X-Product-SHA256 header.
It proves the full path works: challenge, signature, facilitator verification, on-chain settlement, delivery, single-use enforcement and refusal to double-charge. It was run on a test network, so it proves nothing about mainnet throughput or fees, and it is not a claim that the material earns anyone money.
The archive includes an exact-arithmetic calculator that works in USDC micro-units, so no floating point error creeps into pricing decisions. Running it on the bundled example:
revenue 10.000000 USDC 1,000 paid calls at 0.01
total cost 12.100000 USDC
net -2.100000 USDC
contribution/call 0.008000 USDC
break-even 1,263 paid callsThe example is deliberately unprofitable in its first period, because setup cost is real and most first attempts do not clear it. Removing setup cost in a later period yields 2.90 net and a 638-call break-even. Those are the bundled assumptions, not a forecast of your business.