Service · Web development

Automate invoice recognition with AI and OCR.

Automatically read, classify and post purchase invoices to your accounting software or ERP. We build custom invoice processing where standard tools fall short: industry-specific, with deep integrations, and ready for Peppol and the EU e-invoicing mandate in 2030.

OCR & LLMPeppol / UBLERP integration3-way matchGDPR & audit

For when standard scanners don't go far enough.

Klippa, Basecone, AutoInvoice, FactuurSturen and Visma Scanner do a fine job on standard PDF invoices, and we don't build a copy of them. Our clients come to us when those tools hit a wall: an industry-specific invoice format the standard parser can't read, an ERP integration that needs to go just a little deeper, or an approval flow with multiple cost centres and budget categories that fits no SaaS form. Often it's a combination of factors too: an organisation that doesn't just want to recognise invoices, but also match them directly to purchase orders, contracts and framework agreements held in its own system.

We build custom invoice recognition with modern OCR (Tesseract, PaddleOCR, Google Vision) and multimodal models such as Anthropic Vision and GPT-4o Vision that truly understand the nuances of an invoice. The result is a processing pipeline that fits into your ERP, not the other way round, and that grows with new supplier formats, new budget structures or a future move to Peppol-only e-invoicing. Request this for your accountancy firm, for a procurement flow in a large organisation, or as part of broader document workflow automation.

Our clients are SMEs with fifty to five hundred purchase invoices per month, accountancy and bookkeeping firms processing client administrations, and large organisations with formal procurement flows. What they share is a breaking point: the volume or complexity is too high for manual entry, but too specific for an out-of-the-box scanner.

Three scenarios where custom makes the difference.

Not every organisation needs the same thing. In our first conversation we advise which approach fits your invoice volume and system landscape.

Compact project · fixed sprint budget

SME invoice processing with deep accounting integration

For SMEs that manually book fifty to five hundred purchase invoices every month. We build an ingestion flow where invoices arrive by email, drop folder or Peppol, are read automatically by a combination of OCR and a language model, and then land directly as a journal entry in Exact, Twinfield or AFAS. The approver gets a notification, approves with one click, and the invoice is ready for payment.

OCR + LLM extractionExact / Twinfield / AFASApproval flowMailbox ingestion
Mid-sized project · fixed sprint budget

Accountancy firm: processing client invoices in batches

Accountancy and bookkeeping firms often work for dozens of clients, each with their own accounting package and their own administration. We build a central processing pipeline that knows the right chart of accounts for each client, automatically suggests the right general ledger account and cost centre, and only brings in a human when in doubt. The LLM learns from previous bookings: the same supplier on the same account, automatically.

Multi-tenantGeneral ledger suggestionsHuman-in-the-loopAudit trail
Larger project · fixed sprint budget

Procurement flow with 3-way match and Peppol

For large organisations with a formal procurement flow: an incoming invoice is automatically matched to the purchase order and the goods receipt (3-way match), with escalation to the right budget holder when there are discrepancies. Full Peppol BIS support for e-invoices, native UBL processing, and integration with SAP, Oracle or Microsoft Dynamics. An audit trail for every step, ready for your auditor's checks.

3-way matchPeppol BISNative UBLSAP / Dynamics

What you get at the end.

A working processing pipeline in production, plus the tools to adjust and extend it yourself.

  • Live processing pipelineProduction and staging environments, hosted in your own cloud (GCP / AWS / Azure) or with us. Inbound channels for email, upload, Peppol and API ingestion.
  • Management dashboardOne screen for your finance team: invoices in progress, escalations, turnaround times per supplier, and patterns that need attention.
  • Full codebase + documentationSource code in your repository, with an architecture description, deployment instructions and an incident runbook. No vendor lock-in.
  • GDPR and audit reportingDPIA template, data catalogue, retention periods set per document type, and an audit log that shows, for each invoice, who changed or approved what.
  • Training for finance + ITTwo sessions for key users (how to handle escalations, how to adjust booking rules) and a technical session for your IT team on management and maintenance.
  • Support contract (optional)Ongoing maintenance, model updates, new supplier formats and integration extensions. Fixed monthly price, four response-time levels.

The technical stack behind invoice recognition.

We aren't tied to a single vendor or model. For each project we choose the combination that fits your invoice mix, your cloud environment and your budget, and we explain why.

Layer 1 · Document extraction

OCR and multimodal vision models

For clean PDFs and e-invoices we work directly with the structured data. For scans and photos we combine classic OCR (Tesseract and PaddleOCR for local processing, Google Vision when the cloud is acceptable) with multimodal models such as Anthropic Vision and GPT-4o Vision. These models read an invoice as a whole: they understand that a number in the top right is the invoice number, even if the label says "Invoice #" or "Faktura Nr.". For sensitive documents we run local models in your own environment, so invoice data never leaves your data centre.

TesseractPaddleOCRGoogle VisionAnthropic VisionGPT-4o Vision
Layer 2 · Data validation

LLM-based extraction and validation

The raw text goes to a language model that pulls out the fields you need: invoice number, invoice date, due date, VAT amounts per rate, cost centre, project code, supplier details, IBAN. We build business-rule validation around it: does the VAT total add up, does the invoice date fall within the financial year, do we know this supplier? If there is any doubt, the invoice goes to a human reviewer with the doubtful points already flagged. It is a form of human-in-the-loop that converts quickly to fully automated once the model knows the patterns.

Field extractionBusiness rulesVAT validationIBAN check
Layer 3 · Integration

Accounting, ERP and Peppol flow

The extracted data is then pushed to your system. For SMEs that is usually Exact Online, Twinfield, AFAS or Visma. For accountants, often Yuki or Snelstart per client. For enterprise, SAP, Oracle NetSuite, Microsoft Dynamics or Unit4. We process incoming Peppol invoices natively as UBL, and for outgoing invoices we connect to a Peppol Access Point. For specific integrations not covered here, we look at the existing API and build an adapter. You can read more on our page about smart API integrations.

Exact / Twinfield / AFASVisma / YukiSAP / DynamicsPeppol Access Point

When custom is the right choice.

Four signals we see in organisations that have outgrown their standard scanning tool. If you recognise one of them, we would be glad to talk further.

Industry format

Your suppliers use a different invoice format

Suppliers in construction, healthcare or agriculture send invoices with lines and codes that standard scanning tools cannot separate out. A custom extraction model that understands your invoice patterns suddenly delivers clean data.

Deep ERP integration

Your accounting integration needs to do a bit more

Standard tools write to a single field in Exact or Twinfield. You have cost centre, project and VAT code logic that differs per invoice, and you want it filled in automatically before posting.

Volume & scale

The volume does not justify manual work

At fifty invoices a month, manual entry still works. At five hundred or more, it becomes a full-time job. A custom processing pipeline pays for itself there, especially if it is allowed to reach deeper into your processes.

Peppol & 2030

You are preparing for mandatory e-invoicing

The EU will make business-to-business e-invoicing mandatory via Peppol-style standards from 2030. If you build a platform now that can handle both PDF and UBL, you will not have to switch in a hurry later.

How an invoice recognition project works.

1

Introduction and data sample

We review a representative set of your invoices (typically one to two hundred documents across your main suppliers) and your current posting process. From that comes an initial estimate of extraction quality per field, which integrations we need, and where the biggest time savings are. In this phase we are honest about what we can and cannot solve for you. Sometimes a standard tool supplemented with a small custom connector is more than enough, and our clients would rather hear that up front than halfway through a project.

2

Architecture and proof of concept

A short design sprint with your finance and IT teams. We choose the right OCR stack (Tesseract, PaddleOCR or Google Vision), decide which LLM does the extraction work, and set up a proof of concept your team can use to test the first batch.

3

Building in sprints

A working build every two weeks: more supplier formats, integration with your accounting system, approval flow, audit log. You test continuously alongside us, and finance joins for the UX decisions.

4

Pilot and phased rollout

We start with one department or one set of clients, run in parallel with the old process for a while, and expand once extraction quality and approval speed are up to standard.

5

Further development and maintenance

New suppliers, new integrations, model updates and expansion into areas such as computer vision applications for receipts or contract analysis. Many clients expand to full procurement automation in year two.

Frequently asked questions.

What clients usually want to know before we start.

Do you replace tools such as Klippa, Basecone or AutoInvoice?
No. For standard PDF invoice processing, those tools do a good job and are far cheaper than custom development. We come into the picture when you hit their limits: an industry-specific invoice format they can't read, an ERP integration that needs to go just a little deeper, or an approval flow that doesn't fit into any SaaS form. Our solutions often run alongside an existing tool, not instead of it.
What is the difference between classic OCR and AI-based extraction?
Classic OCR (Tesseract, PaddleOCR) converts pixels into text. That works fine for clean PDFs but struggles with stained scans, unusual layouts or handwritten notes. Multimodal models such as Anthropic Vision and GPT-4o Vision read an invoice as a whole in one go: they understand that a number at the top right is the invoice number and the amount at the bottom is the total, even in unusual formats. We combine both: OCR for the groundwork, a language model for interpretation and validation.
How do you support Peppol and UBL?
Peppol is a European e-invoicing network and UBL is the underlying data format. We build in native UBL processing: invoices that arrive via Peppol are ingested directly as structured data, with no OCR detour. For sending, we connect to a Peppol Access Point. That matters given the EU e-invoicing obligation that takes effect in 2030: a platform that can already handle both PDF and UBL won't have to migrate in a rush later.
Which accounting and ERP systems do you integrate with?
We have experience with Exact Online, Twinfield, AFAS, Visma, Yuki and Snelstart, and on the ERP side SAP, Microsoft Dynamics, Oracle NetSuite and Unit4. For every integration we look at the official API offering and at what the system can and can't handle in practice: some accounting packages have tight APIs, others only accept CSV imports. For topics beyond direct accounting integration, see also our smart API integrations.
What about GDPR, data location and audit trail?
Invoices contain personal data (names, bank details, sometimes BSN numbers for freelancers). We draw up a DPIA for every project, encrypt data in transit and at rest, and host in European data centres with providers that have a data processing agreement. An audit log is standard: for each invoice you can see who changed, viewed or approved what. Retention periods can be set per document type, in line with the seven-year retention obligation for accounting records and the GDPR data minimisation requirement.
What determines the cost of a project like this?
Three factors: how many different invoice formats you process (a set of ten suppliers is much faster to model than a hundred), how many integrations we need (one accounting package or four ERPs plus an approval platform), and the complexity of your approval flow. After the introductory meeting we give a well-founded estimate per sprint: not a fixed total price based on a gut feeling, but a transparent budget per phase.
Can we start with a proof of concept?
Yes, and we often recommend it. In a short initial phase we train the extraction model on a sample of your own invoices, connect one accounting package, and you receive a report stating exactly what the extraction quality is per field and per supplier. Only then do you decide whether to continue with the full build: no surprises afterwards. Many clients also use this POC to secure internal budget for the full rollout. Also read about our approach to custom LLM integrations that underpin this.

Talk to us about your invoice processing.

A no-obligation introductory call of half an hour. Send a few sample invoices along and we can already show you, during that conversation, where standard tools reach their limits — and where custom development makes the difference.

Edit content