Solutions / Reconciliation

Reconciliation & ledger sync

Every transaction across your payment providers, bank accounts and ledger is matched automatically each day — and the only thing that reaches a person is the handful that genuinely does not line up.

Finance System integrations Finance · E-commerce
100%
of entries reconciled daily
Same day
a gap appears, not at close
Exceptions only
reach a human queue
Who it is for

Finance and fintech ops teams closing the books monthly across more than one payment provider, where the reconciliation spreadsheet has quietly become a full-time job.

Short answer

Transactions from your PSPs, bank accounts and ledger are matched every day, automatically. 100% of entries are reconciled daily and a discrepancy surfaces the same day it appears rather than three weeks later during close — and only genuine exceptions land in someone's queue.

The problem

You find out about February's gap in March, when nobody remembers February.

01

Close is a week of matching rows

Exports from two PSPs, three bank accounts and the ledger, joined by hand in a spreadsheet each month. It works, it takes days, and it is one departure away from nobody knowing how.

02

Discrepancies age before anyone sees them

A missing settlement or a double charge found five weeks later is an archaeology project. Found the same day, it is a two-minute email to the provider.

03

Fees and refunds break naive matching

Amounts rarely match to the cent — provider fees, partial refunds, chargebacks, FX, batched payouts covering forty orders. Rule-based matching handles the easy 70% and dumps the rest on a person.

How it works

From trigger to result, step by step.

01

Pull every source on a schedule

PSP settlement reports, bank statements over API or file feed, and ledger entries — normalised into one shape with currency, date, reference and counterparty made comparable.

02

Match exactly first

Reference, amount and date. This clears the large majority and costs nothing, and it means the harder logic only ever sees what is genuinely ambiguous.

03

Then match what does not match cleanly

Batched payouts split back into their orders, fees and FX differences accounted for, partial refunds and chargebacks tied to the original charge, near-duplicate references resolved.

04

Raise exceptions with context

Whatever remains becomes a ticket carrying both sides of the comparison, the amount and direction of the gap, and its probable cause — not a highlighted spreadsheet row.

05

Write results back and keep the trail

Matches are posted to the ledger with their evidence. Every decision is logged and re-runnable, so an auditor can retrace how any entry was matched, months later.

Before / after

What changes on the ground.

Today, by hand
×Close takes a week of manual row matching
×A gap from February is discovered in March
×Fees, FX and batched payouts break the match
×The process lives in one person's spreadsheet
With the automation running
›100% of entries reconciled every day
›A discrepancy is visible the day it appears
›Batched payouts, fees and refunds matched properly
›An audit trail that explains every match
What you get

Delivered, not demoed.

›Connections to your PSPs, banks and ledger with normalised transaction data
›A daily matching run with exact and fuzzy passes
›An exception queue with both sides and the probable cause
›A re-runnable audit trail for every matched entry
›Documentation and a handover session — the system is yours
Built with

We build in your stack rather than moving you onto ours. The list below is what this solution most often connects to — other systems are a scoping question, not a blocker.

Stripe PayPal Bank APIs / statement feeds QuickBooks Xero 1C PostgreSQL Python n8n
Time to production6–8 weeks
Build priceFixed quote
First stepFree mini-audit
Honest limits

When this is not the right solution.

·If your ledger entries carry no usable reference to the payment, matching has nothing to work with. Fixing what gets written at the point of sale comes first, and the audit will show you exactly where it breaks.
·One payment provider and thirty transactions a month reconciles fine by hand. We will tell you that rather than build around it.
·This is not an accounting system and does not replace your accountant. It matches, flags and explains — the judgement calls and the postings policy stay yours.

Questions we get about this one

It raises an exception with both sides shown, the size and direction of the gap, and its best guess at the cause — a provider fee, a partial refund, a timing difference across a weekend. It never invents a match to make the numbers look clean, because a wrong match is far more expensive than an open exception.

Bring us the process that hurts.

The mini-audit is free: we take your version of this process apart and tell you plainly whether automating it pays. If it does, you get a scope and a fixed price.