Bank statements are downloaded as XML or CSV and matched to invoices by hand, line by line, and unclear references are chased with the client. AI matches lines by reference, amount and history, prepares payment entries in SAP Business One or Sage and lists unmatched lines separately. The accountant approves.
Built for ISO 27001NIS2 DORA
Every client of an accounting firm has at least one bank account, often two or three at different banks. At the start of the month the accountant downloads the bank statements as SEPA XML or CSV files, imports them into SAP Business One or Sage and starts the work the import does not do: assigning each line to a sales invoice the client was paid for, a supplier invoice the client paid, salaries, payroll taxes, tax authority payments or bank fees. The reference often says only "for services" or contains a number with a typo, an amount covers three invoices, and a partial payment matches nothing. Then the client is asked, the firm waits for an answer, and an unreconciled statement gets finished only a day before the VAT return is due.
The solution reconciles what can be reconciled and openly shows what cannot. Statement XML and CSV files are taken from the client's SharePoint folder or an Outlook e-mail, and the lines are compared with unpaid sales and purchase invoices in SAP Business One or Sage: by the invoice number in the reference, the payer or payee code and the amount, and where the reference is empty, by that client's history of how the same payer and amount were matched before. Payments to the tax authority and the social security authority, salaries and bank fees are recognised by payee and assigned to the corresponding accounts. Payments covering several invoices in one amount are split, and partial payments are flagged. The result is payment entries in SAP Business One or Sage with the status "awaiting approval" and a separate list of unmatched lines with suggestions on what to ask the client. The accountant approves, moves or rejects; no payment is matched without them. We deploy private AI in MDP CLOUD's GPU cloud in Lithuania (EU): statements that show clients' and their partners' transactions do not leave our data centre, the models do not learn from your data, and every match and every accountant decision is logged in line with GDPR.
Four steps from input to a record in your system, with human approval wherever there is risk.
Bank statement files in XML or CSV format are taken from the client's SharePoint folder or an Outlook e-mail and assigned to the client and their bank account by account number. Formats from different banks are normalised, and an already imported period is recognised so that lines are not duplicated.
Each line is compared with unpaid invoices in SAP Business One or Sage by the number in the reference, the payer code and the amount, and where the reference is unclear, by that client's history. Lines for tax authority and social security payments, salaries and bank fees are recognised by payee. Several invoices paid with one payment are split.
Matched lines appear in SAP Business One or Sage as payment entries with the status "awaiting approval" and a link to the statement line. Unmatched lines go on a list with a suggestion of what to ask the client. The accountant reviews, approves, moves a line to another invoice or rejects it.
Partial payments, new payers with no history and amounts that match several invoices equally well are not matched but left for the accountant with options. Every month we review how many matches were corrected and tune the rules for each client with the chief accountant.
Matched lines are created in SAP Business One through an API or import file and in Sage through import as unapproved payment entries with a link to the statement line.
SEPA XML (ISO 20022) and CSV statements from different banks are normalised and processed in the same way, and imported periods are tracked so that lines are not duplicated.
Every morning in Teams the accountant sees each client's unmatched lines with suggested options and a prepared question for the client, which the accountant sends themselves.
Every match and accountant decision is logged with a time and a user. Training for accountants and tuning the rules during the first month are part of the implementation.
Private AI in the MDP CLOUD cloud + SAP Business One and Sage import. We choose the platform by the task and by where your data may be located. Read more about this service and its limits on the service page.
No. SAP Business One or Sage receives only payment entries with the status "awaiting approval", and the accountant posts them in one step after reviewing the match. Doubtful lines do not become entries at all, they stay on the list with options.
By the client's history in SAP Business One or Sage: what that payer and amount were matched to before and which of the client's invoices are currently unpaid. If there is no history or several invoices fit equally well, the line stays unmatched and the accountant gets a prepared question for the client.
Yes: with SAP Business One through an API or import file, with Sage through import, and with QuickBooks Online, Xero and NetSuite through import or export files, depending on the version. We take statements from files you download from the bank and do not connect to bank systems directly. We have no official partnerships with accounting software vendors; we work with their public interfaces.
After the 10th, accountants write to clients about missing invoices, bank statements and expense reports.
Clients ask in Outlook every day how much VAT and payroll tax to pay, by when and where their invoice is, and the accountant opens SAP Business…
Clients send invoices as phone photos, supplier PDFs and forwarded e-mails, and the accountant keys them in by hand.
A question about VAT deduction or a personal income tax relief costs half an hour of searching tax authority rulings, and the firm's procedures…
This service is delivered by specialists with valid international certifications. Their expertise is confirmed by independent organisations, not by us.
AI systems are deployed with protection against data leakage and model abuse.
The AI security programme and its controls are managed following the ISACA methodology.
AI risks are identified and managed like any other IT risk.
We reply within 1 business day
