Bank reconciliation is an accounting control that compares a company’s internal cash records against the bank statement for the same period, so any transaction that appears in one but not the other can be found, explained, and corrected before the books are closed.
What bank reconciliation verifies
On paper, the cash balance in a company’s general ledger should equal the balance its bank reports. In practice, the two rarely agree on any given day.
Payments clear on the bank’s timeline rather than the company’s, deposits post after they are recorded internally, and the bank applies its own fees, interest, and adjustments that finance has not booked yet.
Reconciling the two answers a single question: is the cash we say we have the cash we actually have? Once the book balance and the bank balance are adjusted for these known differences and still fail to agree, the remaining gap points to something worth investigating, a duplicate payment, a missing entry, or, occasionally, fraud.
Why it is a core financial control
Cash is the account most exposed to error and misappropriation, which is why bank reconciliation is treated as a control, not just bookkeeping.
It is usually one of the first tasks in the month-end close, and internal-control frameworks such as SOX expect cash to be reconciled and independently reviewed on a fixed cadence rather than left until year-end.
The requirement grows with complexity.
A business running hundreds of accounts across currencies and subsidiaries reconciles each one on its own, then rolls the results into the wider financial close.
The more accounts and transactions involved, the more the process depends on consistent rules instead of one-off spreadsheet checks.
How it works
Most teams follow the same sequence, adjusting each side until the two balances meet:
- Set the starting balances: Record the ending bank balance from the statement and the ending book balance from the cash ledger for the same date.
- Adjust the bank side: Add deposits in transit and subtract outstanding checks the bank has not processed yet.
- Adjust the book side: Record items the bank already knows about but the ledger does not, such as service fees, interest, and NSF checks.
- Compare the adjusted totals: The corrected bank balance and corrected book balance should now match.
- Resolve and approve: Investigate any remaining difference, post the adjusting journal entries, and have a reviewer sign off.
Example
Suppose a company’s cash ledger shows a closing balance of $89,300, while the bank statement reports $90,500, a $1,200 difference to reconcile.
On the bank side, a $2,000 deposit in transit arrived on the last day of the month but posts the next business day, and $3,500 in outstanding checks to suppliers have not cleared. On the book side, the bank charged a $300 service fee that finance has not yet recorded.
After adjusting the bank balance for the deposit and the uncleared checks, and the book balance for the fee, both sides land on the same adjusted balance of $89,000. The reconciliation ties out, the fee is posted as a journal entry, and the account is ready for review.
What causes bank statement discrepancies
Most differences fall into two buckets: timing differences that resolve on their own, and true errors that need a correcting entry.
- Deposits in transit: cash recorded internally but not yet posted by the bank.
- Outstanding checks: payments issued that have not cleared.
- Bank fees and interest: amounts the bank applies before finance records them.
- NSF checks: deposits reversed because the payer’s account lacked funds.
- Recording errors: transposed digits, duplicate entries, or postings to the wrong account.
Manual vs. automated bank reconciliation
As transaction volume climbs, hand-matching in spreadsheets becomes a bottleneck. Here is how the two approaches compare:
| Dimension | Manual (spreadsheets) | Automated (platform) |
|---|---|---|
| Data collection | Statements and exports pasted in by hand | Bank, card, and ERP data ingested via connectors |
| Matching | Line by line, every close | Rule-based auto-matching; only exceptions surface |
| Auditability | Easy to miss, hard to trace | Consistent rules with a full audit trail |
| Scale | Breaks down as accounts grow | Handles many accounts and high volume |
| Close speed | Slower, a recurring bottleneck | Faster and repeatable |
How Simetrik automates bank reconciliation
Simetrik is an AI-powered financial control platform built for this work. Its pre-built connectors pull data from banks, payment processors, card networks, and ERPs such as SAP, Oracle, and NetSuite, normalizing formats automatically even when a provider changes its file structure.
From there, a deterministic matching engine clears the transactions that tie out under the rules your team defines, while an AI layer suggests new matching rules and sharpens accuracy over time.
Whatever does not match lands in an exceptions hub with severity and root-cause context, so analysts resolve real issues instead of ticking off matches. Every step is logged, so the reconciled result is audit-ready and flows straight into journal entries and the wider close.