The morning task that never ends
At many Indonesian rural banks (BPR and BPRS), the operations day starts the same way. Files are downloaded from the ATM switching network, bill payment reports are opened, statements from partner banks are pulled, and everything is matched against core banking data in a spreadsheet. Row by row.
With low volumes, this still works. The problem is that channels keep multiplying: shared ATM networks, bill payments, interbank transfers, agent deposits, QRIS. Each channel brings its own file format, its own delivery schedule, and its own way of recording a transaction. The spreadsheet that used to be enough has become a pile of files only one or two people truly understand.
Three costs nobody records
The cost of manual reconciliation is rarely calculated because it never appears as a single line on the financial statements. It is spread across three places.
1. Staff hours. A simple illustration: two staff members spend three hours a day matching transactions. Over a month of 22 working days, that is 132 hours, almost one full-time employee, spent only on confirming that two systems agree. Month-end overtime is extra.
2. Discrepancies found too late. A cash withdrawal that succeeded at the switch but failed to post in core banking. A transfer debited twice because it was resent after a connection dropped. A bill payment stuck in a suspense account. If reconciliation finishes two or three days later, these discrepancies are found two or three days later too, by which time funds may have moved and tracing them is far harder.
3. Customer trust. To a customer, a debited balance with no cash from the machine is not a technical issue. It is a reason to switch banks. Rural banks compete on closeness and trust, and one slow complaint travels faster than any promotion.
Why it is more urgent now
Indonesia's Financial Services Authority (OJK) made digital acceleration one of the pillars of its 2024-2027 Roadmap for the Development and Strengthening of the BPR and BPRS Industry. The implementing regulation is already out: POJK Number 34 of 2025 on Information Technology Implementation by BPR and BPR Syariah, which stresses comprehensive IT governance and risk management.
The question supervisors and internal auditors will ask more often is no longer just "do the numbers match," but "who matched them, when, under which rule, and who approved the correction." A spreadsheet struggles to answer that cleanly.
What healthy reconciliation looks like
Automated reconciliation is not simply moving a spreadsheet onto a nicer screen. Five things set it apart.
- Every source in one place. Switching files, core banking, partner banks, and billers are loaded through per-channel templates, so format differences are no longer handled by hand.
- Matching runs on clear rules. Matching keys, time tolerances, and product-specific rules are defined once and applied to every transaction.
- Discrepancies go straight to a queue. Unmatched transactions do not get buried on row ten thousand. They appear in an exception queue and can be rematched as soon as late data arrives.
- Corrections pass four eyes. Every adjustment is created by a maker, reviewed by a checker, and approved by an approver. Nobody can create and approve their own correction.
- Every trace is kept. Who changed what, and when, is recorded automatically and ready for audit.
SIRECONS: built to a regional bank standard
SIRECONS is XETUP's reconciliation and settlement system. It is currently being implemented at an Indonesian regional development bank, matching transactions from multiple payment channels against core banking.
What matters for rural banks:
- Runs on the bank's own servers. Customer transaction data never leaves the bank's environment.
- Sign in with existing accounts. SIRECONS can connect to the bank's user directory, so there are no separate passwords to manage.
- Per-channel templates. New channels are added through templates, not by rebuilding the system.
- Fast enough for growing volume. In internal testing, the matching engine processes tens of millions of rows per minute, far above a rural bank's daily volume.
- Built-in approval flow and audit trail, following the four-eyes principle common in banking.
For rural banks, we offer an edition sized to their scale and leaner channel count, without lowering the security and control standard.
See it working in 45 minutes
The fastest way to judge whether your bank's reconciliation is worth automating is to watch it run. In a SIRECONS demo, we walk through the full flow: loading files from several channels, automatic matching, exception handling, tiered approval, and the audit trail.
If you like, bring a sample file format from the one channel that causes the most trouble (masked data is fine), so the demo uses data shaped the way your team sees it every day.