Payment Chaos

The measured day, outcome by outcome.

On one anonymised merchant day, Reconora reconciled 114,065 deposit transactions against provider notifications and statements: 75.9% of turnover was matched, 10.1% had nothing to reconcile against, and 809 payments went to the discrepancy queue with a reason. This page walks through that day; it is one observation, not a benchmark.

Reconora Research · One anonymised merchant day. The merchant, its providers and the date are not named. The Payment Chaos report describes the messages of the full day; this page follows the reconciliation scope — the merchant’s deposit log reconciled with notifications and statements. How every number is produced

What went in

The merchant exported one day of its orchestrator log, with the provider messages it records, and the providers’ statements. The logs were anonymised on an ordinary laptop at the merchant’s site in 40.3 minutes and the statements in 2.9 minutes; the whole run to the first result took 63.1 minutes of machine time. The reconciliation scope held 114,065 deposit transactions.

Matched: 75.9% of turnover

A matched payment is one where the merchant’s log, the provider’s notification and the provider’s statement agree on the payment, its amount and its outcome. That share of turnover could be booked with confidence. A spreadsheet lookup on the same day “found” 75.5% of log IDs in the statements, but only 74.4% of lines truly agreed: 690 lines were found and still wrong. Spreadsheet vs Reconora

Differs in facts: 1.3% of turnover

Differs in facts means the sources disagree on whether a payment happened or how it ended. On the measured day that included 320 payments declined in the log and successful in the statement — players charged but not credited — and 61 payments the other way round. A further 317 payments appeared in a statement with no record in the merchant’s log, and 51 were missing from a statement that should have held them. Reversed outcome

Differs in money: 0.1% of turnover

Differs in money means the sources agree on the payment and its outcome but not on the amount. The share is small on this day, and every such payment carries the stated difference, so it can be reviewed and, where provable, raised with the provider instead of disappearing into a rounding line.

Nothing to reconcile against: 10.1% of turnover

Nothing to reconcile against is volume with no document to check it: 26 providers sent no statement at all, some statements were missing for a currency, and some held only part of the provider’s payments for the day. At one provider the log showed 1,207 successful payments while its statement listed 314 — 26%. Booking the difference as a debt would have been wrong; requesting the full statement was the right step. Nothing to reconcile against

Awaiting a statement: 12.6% of turnover

Awaiting a statement is not a discrepancy: the provider’s statement does not cover that hour yet. Payments near midnight are the typical case — 464 lines sat at the day edge and wait for the neighbouring day’s statement. Separately, 14,191 transactions had no final status from the provider at the end of the day; they are tracked until one arrives. Day edge

What came out

The day produced 809 payments in the discrepancy queue, each with a reason and the address of its source rows, and 11 claim candidates covering 128 payments, grouped by provider and problem with their evidence. Blind spots were listed by provider and currency, so the merchant knew which statements to request first. Discrepancies and claims

What one day can and cannot tell you

One day shows what happens, with exact counts. It does not show how often the same patterns occur at other merchants, in other verticals or in other seasons. Your providers, currencies and statements set your own mix — which is why the pilot measures one day of your own data.

Try it on one day of your data

One day of your data, reconciled, and the report stays with you whether or not you continue.

By Reconora Research · Updated