AI infrastructure for tax advisories in Germany.
AI infrastructure for tax advisories in Germany automates recurring tasks such as voucher capture, matching, and filing, without breaching tax advisors' professional conduct rules or GoBD-compliant retention. NexPatch connects systems such as DATEV and your existing document management, so AI agents pre-sort vouchers and practice staff handle the professional review - instead of capturing data by hand.
Professional rules and GoBD: requirements every solution in the practice must meet
AI in a tax practice does not operate in a legal vacuum. Three requirements are binding for every solution, regardless of which specific task is being automated. NexPatch evaluates them with the practice before the project begins, rather than accounting for them after deployment.
Confidentiality under §57 StBerG and §203 StGB
Tax advisors are bound to statutory confidentiality under §57 StBerG; breaches can be punishable under §203 StGB. An AI solution may therefore only assist with pre-sorting and suggestions; professional verification and sign-off remain strictly with the practice. Client data is processed exclusively on systems controlled by the practice, with zero transfer to third parties. Similar requirements apply to legal secret holders in law firms, as shown on our AI for Law Firms page.
External service providers under §62a StBerG
When a firm engages an IT or AI vendor, §62a StBerG requires more than a standard GDPR Data Processing Agreement (DPA). The provider must be vetted, bound to professional secrecy in text form, and instructed on the criminal liability consequences. They may only access what is strictly necessary. When AI is operated locally, the provider has zero ongoing access to client data during operations.
Audit-proof retention under GoBD
Receipts and booking transactions must be stored immutably, completely, and traceably throughout the statutory retention periods under §147 AO, as mandated by GoBD. An AI solution must therefore log every processing step. If an AI agent alters a categorization, the original record remains preserved in audit logs. Storage occurs in the firm's certified document management system, not in an unverified external tool.
Locally operated: client data never leaves the firm
Many AI tools process input on servers outside the EU. For a tax advisory firm, this presents serious risks under §203 StGB and §62a StBerG that cannot be reliably remedied by contract alone. NexPatch takes the opposite path: language models run on local hardware or dedicated German servers, ensuring client data remains strictly on-premise.
The difference from a pure software purchase lies in ongoing operations. NexPatch provisions the private AI infrastructure, builds the AI agents, and retains full operational accountability post-deployment: updates, monitoring, adapting to new document formats, and providing incident support. The firm receives a running system rather than an unmanaged maintenance burden. Learn more on our operations page.
Typical workflow: from voucher intake to booking
In practice, automated voucher capture runs through several traceable steps. A voucher arrives - as a scan, an email attachment, or an upload from the client. An AI agent reads the voucher, matches it to a client, account, and booking circle, and files it in the document management system. Vouchers with a clear, recurring pattern are proposed directly for booking; vouchers with anomalies or unusual details are flagged for manual review. Only after the practice signs off does a booking actually enter DATEV. That final step stays with a person in every case, regardless of how confidently the AI agent assessed the voucher.
pages.industriesTaxAdvisors.secDmsH2Title
pages.industriesTaxAdvisors.secDmsH2P1
Connecting DATEV and document management
Official DATEV interfaces
DATEV is the central core accounting application for bookkeeping and financial statements in most German tax firms. NexPatch connects AI agents exclusively via official DATEV interfaces. This keeps solutions robust across DATEV updates, preserves access permissions, and maintains transparent data flows. Staff continue working in DATEV, with AI enhancing the workflow rather than replacing it.
| Interface | How AI utilizes it |
|---|---|
| Belegbilderservice | Transfer pre-sorted receipts directly into DATEV Unternehmen online |
| Buchungsdatenservice | Transfer reviewed booking proposals as batch booking files |
| Rechnungsdatenservice | Pass structured invoice data, including e-invoices |
| DATEVconnect | Query master data and connect custom on-premise integrations |
Document management system
The practice's own document management system handles GoBD-compliant filing. Incoming vouchers - by email, scan, or client upload - are captured, classified, and filed automatically in a way that meets retention requirements and stays findable for later review. Follow-up questions to clients, for example about missing or unclear vouchers, can be documented traceably through the same structure, instead of getting lost across separate email threads.
Use cases in the tax practice
Voucher capture is the most common entry point. Once the local AI infrastructure is live, further workflows can be layered onto the same platform without procuring additional disparate systems.
Automating voucher capture
Automating voucher capture is the primary entry point for AI in a tax practice. The following model calculation uses disclosed assumptions, not an anecdotal customer claim.
| Assumption | Value |
|---|---|
| Voucher volume per month at the example practice | 3,000 vouchers |
| Time per voucher for manual capture and matching | 3 minutes |
| Calculated total time per month without automation | about 150 hours |
| Share of vouchers with a clear, recurring pattern per model assumption | about half |
| Calculated time saved for that share, per month | about 75 hours |
For that share, AI handles pre-sorting and matching, while the practice reviews and signs off. The actual saving depends on the voucher mix, the client base, and how far the practice has already digitized, and is reviewed individually for each practice before it becomes a firm planning figure for the practice team.
Calculate the exact savings for your firm using our ROI calculator.
Requesting missing receipts from clients
An AI agent automatically detects missing or incomplete receipts for a booking period and drafts clarification requests to the client. The firm reviews the communication, and client replies are automatically linked to the transaction in the DMS instead of being scattered across mailboxes.
Processing electronic invoices
With the staggered B2B e-invoicing mandate in Germany, structured invoices such as XRechnung and ZUGFeRD are rapidly increasing. An AI agent extracts structured data, reconciles it against image receipts and payments, and transmits it to DATEV via the Rechnungsdatenservice.
Searching internal firm knowledge
Internal guidelines, client records, and institutional know-how often sit scattered across folders, emails, and notes. A private, local RAG retrieval system answers staff questions using internal documentation with verified citations, while all data remains strictly on-premise.
How a project starts at a practice
An initial internal pilot using real vouchers from the practice stands within 72 hours, and productive rollout into day-to-day operations follows in 4 to 10 weeks, depending on the system landscape. Which AI agents come into play is shown on the AI agents overview. Technical details are described on the private AI infrastructure page. Learn more about how long an AI rollout takes in our knowledge base, and read how NexPatch manages systems after go-live on our operations page.
In an initial consultation, we analyze your voucher intake, DATEV environment, and document management system to identify high-impact automation potential.
Download DATEV Interfaces Fact Sheet (PDF)Frequently asked questions
In an initial consultation, we analyze your voucher intake, DATEV environment, and document management system to identify high-impact automation potential.
Download DATEV Interfaces Fact Sheet (PDF)