The verdict in three sentences
When two accountants spend 3 days a month ticking off bank statements for a company with 50 million in revenue, the problem is not skill but the lack of daily bank feeds wired to matching rules. A well-configured connector auto-matches 85 to 95% of transactions and leaves the team only the genuine exceptions. The project costs roughly 9,000 to 22,000 USD (8,000 to 20,000 EUR, 2026 market range), goes live in 6 to 10 weeks and saves 30 to 40 person-days a year.
Where the time goes today and what auto-matching changes
Manual matching almost always follows the same pattern: download statements at month end, import them into the accounting system, search invoice by invoice, then chase customers about unidentified receipts. The table compares a typical setup before and after automation.
| Monthly task | Before (manual) | After (daily feeds and rules) | Monthly gain |
|---|---|---|---|
| Retrieving statements | 2 h, downloads from 6 banks | 0 h, daily API or host-to-host feeds | 2 h |
| Matching customer receipts | 14 h | 2 h, reviewing exceptions | 12 h |
| Matching supplier payments | 8 h | 1 h | 7 h |
| Bank fees and charges | 3 h | 0.5 h, rules on narrative | 2.5 h |
| Researching unidentified receipts | 12 h | 4 h, suggestions by amount and reference | 8 h |
| Month-end reconciliation | 9 h | 1.5 h | 7.5 h |
| Total | 48 h, or 6 person-days | 9 h, just over 1 person-day | 39 h |
Over a year, that is 39 h × 12, or 468 hours, about 58 person-days. In practice, after heavier closing months and supervision time, a cautious figure is 30 to 40 person-days saved.
What the project includes and how the budget is built
An auto-matching project rests on three layers: bank feed retrieval, the rules engine, and integration with the accounting system or ERP. Price depends on the number of banks, the quality of payment references and the number of entities.
| Component | Content | 2026 market range | Timeline |
|---|---|---|---|
| Bank connector | Open banking API or host-to-host, 3 to 8 banks | 2,200 to 5,500 USD | 2 to 3 weeks |
| Matching rules engine | Amount, invoice reference, payer name, variance tolerance | 3,300 to 7,700 USD | 2 to 3 weeks |
| Accounting or ERP integration | Journal import, customer and supplier updates | 2,200 to 5,500 USD | 1 to 2 weeks |
| Exceptions dashboard | Unmatched items list, assignment | 1,100 to 3,300 USD | 1 week |
| Multi-entity and multi-currency | Consolidation, EUR or GBP accounts | 0 to 3,300 USD as needed | 1 week |
| Bank feed aggregator subscription | Recurring usage fee | 50 to 330 USD a month | Ongoing |
Built-in reconciliation in mainstream accounting packages is often enough for a small business. For a company with six banks, thousands of customer invoices and bulk payments, it hits its limits: a single transfer settling twelve invoices with one narrative stays unmatched. That is where custom rules pay off.
Kolonell scopes this kind of connector from its custom application range and adapts it to the client's banks and ERP: the final figure is set on quote.
Conditions for reaching 90% auto-matching
The auto-match rate depends less on technology than on reference discipline. Three simple actions take the rate from 70% to above 90%.
First, put a structured invoice reference on every invoice and repeat it clearly in the payment block. Second, ask recurring customers to send a remittance advice when they bundle invoices: the engine reads it and splits the amount. Third, set accepted variance tolerances, for example under 5 USD for charges deducted by the customer's bank.
Monthly steering uses a single indicator: the auto-match rate per bank. A falling rate signals a customer who changed habits or a rule to adjust.
Mini case study
Need a professional website?
Kolonell builds websites that attract clients, optimized for the Sénégalese market. Free quote in 2 minutes.
Rachel, finance director of an industrial distributor in Manchester (50 million in revenue, 6 banks, 4,200 customer invoices a month), sees her two accountants spend 3 days each per month on matching, or 6 person-days.
Chosen project: open banking feeds on 6 banks, matching rules and ERP integration, for 16,500 USD plus 165 USD a month. After 3 months, the auto-match rate reaches 91%.
Remaining time: 1.5 person-days a month. Gain: 4.5 days × 12, or 54 person-days a year, of which 35 are actually redeployed to working capital analysis and targeted collections. At a fully loaded cost of 350 USD per person-day, the redeployed gain is 35 × 350, or 12,250 USD a year. Add the 1,980 USD annual subscription: payback is about 18 months, and DSO drops by 4 days thanks to faster follow-up.
FAQ
What is the difference between bank reconciliation and matching?
Reconciliation checks that the ledger balance agrees with the bank balance. Matching links each transaction to the invoice it settles, which accounts for about 80% of the time spent.
Open banking API or host-to-host connection?
Open banking APIs are quick to set up, usually 0 to 50 USD a month per bank. Host-to-host links such as EBICS or SWIFT cost more but cover corporate accounts more reliably.
What auto-match rate is realistic?
Between 85 and 95% for a company whose invoices carry a structured reference. Without reliable references, the rate often stalls around 70%.
How long does the project take?
Plan 6 to 10 weeks, including 2 to 3 weeks for bank onboarding, which is often the critical path.
Will accountants lose control?
No, they validate exceptions and tune the rules. Their time shifts from ticking to analysis, and every automatic match stays traceable and reversible.
Let's scope your project. We map your banks, matching rules and ERP integration with you, with go-live in 6 to 10 weeks. Detailed quote within 48 h. WhatsApp +221 77 596 93 33.
Mohamed Bah
Fondateur, Kolonell
Passionate about digital and entrepreneurship in Africa, Mohamed has been helping Sénégalese businesses with their digital transformation since 2020. Founder of Kolonell, he believes every SME deserves a professional and accessible online présence.
