Skip to content
D W
GoBD-compliant §203 StGB-compliant Q1-Q2

Asset Capitalisation Agent

Identify, capitalise and record assets in the asset register.

Identifies assets requiring capitalisation from invoices, determines acquisition and production costs per HGB Paragraph 255.

Analyse your process
Airbus Volkswagen Shell Renault Evonik Vattenfall Philips KPMG

Check capitalisation requirement by HGB rules, LLM extraction from invoice texts

The agent validates capitalisation requirement and asset class deterministically against HGB Paragraph 255 and the depreciation tables, and uses AI extraction only to identify the asset from unstructured invoice line items.

Outcome: 80 to 90 percent of asset additions captured fully automatically, errors on low-value asset thresholds structurally eliminated, and consistent useful-life assignments across all plants.

67% Rules Engine
11% AI Agent
22% Human

The split between recognition and classification explains why the agent solves 8 out of 9 steps via rules:

29 percent of all audit findings hit fixed assets

Incorrect capitalisation decisions rank among the costliest avoidable errors in fixed asset accounting. Expensing a capital asset depresses profit in the wrong year. Capitalising a recurring cost inflates the balance sheet. In both cases, depreciation, the tax burden and the asset register are wrong for years to come. The Asset Capitalisation Agent prevents exactly this misallocation - not by automating all of accounting, but by structurally decomposing every single capitalisation decision.

Nearly one in three tax audits flags the asset register

The numbers are clear. According to a PwC survey, every second company must pay additional taxes after a tax audit. In 29 percent of affected companies, the findings relate to fixed asset classification (source: PwC Tax Audit Study, 2024). The most common reason: assets were misclassified, ancillary acquisition costs were not correctly allocated, or useful lives were set without traceable justification.

Germany’s Federal Ministry of Finance reports total additional tax revenue from audits in 2024 at EUR 10.9 billion (USD 12 billion) across 140,764 audited businesses. A significant share falls on asset register corrections - because the original capitalisation decision was undocumented or based on flawed assumptions.

For CFOs, that means: the capitalisation decision is not a bookkeeping detail. It is an audit risk with measurable financial consequences.

The capitalisation decision takes seconds - and carries consequences for years

Picture a typical week in accounts payable at a mid-sized company. Forty incoming invoices, six of them above EUR 800 (USD 880) net. Each of those invoices demands the same sequence of decisions: Is this a standalone asset or an ongoing service? Is the amount above the low-value asset threshold? Which asset class? Which useful life per the depreciation table? Do transport costs count as ancillary acquisition costs?

A clerk often makes these decisions in minutes - under time pressure, with incomplete information, sometimes relying on experience rather than the current official depreciation table. The consequences surface years later: during the physical inventory, when assets in the register do not match reality. Or during a tax audit, when the auditor questions the useful life of a special-purpose machine and finds no documentation.

The error source is not a lack of competence. It is structural: every capitalisation decision contains deterministic elements (threshold, asset class, cost calculation) and discretionary elements (useful life for custom equipment, delineation of composite assets). When both merge in a single manual step, traceability disappears.

Nine decision steps separate rules from judgement

The Decision Layer breaks every asset addition into nine discrete steps. Each step has a defined decision-maker: rule engine, AI model, or human.

The identification of the asset from the invoice description uses a language model. An invoice for “assembly and commissioning of production line - Hall 7” contains no EAN number and no asset master record - but enough context to recognise the asset.

The capitalisation check is purely rule-based: net amount above EUR 800 (USD 880), therefore mandatory capitalisation. Below that, low-value asset treatment or pooling applies. The same holds for the cost basis calculation, inventory number assignment and recording in the asset register. These steps are deterministic. They benefit not from human judgement but from consistency and completeness.

Borderline cases are different: assigning a CNC milling machine with custom configuration to the correct asset class can be rule-prepared but not conclusively decided by rules alone. The useful life of a special-purpose machine does not appear in the standard depreciation table. Here the human decides - but on the basis of a pre-structured decision template, not a blank sheet.

The human decides where discretion is required

The Asset Capitalisation Agent operates at Decision Layer tiers 1 and 2. Rule-based steps run automatically with a full audit trail. AI-assisted steps provide suggestions with confidence scores. No step modifies the balance sheet without a documented decision basis.

In practice: when the agent processes a EUR 12,000 (USD 13,200) invoice for a laboratory scale, the capitalisation check runs rule-based. The asset class “technical plant and machinery” is assigned by rule. The useful life of ten years comes from the official depreciation table, source reference documented. Ancillary acquisition costs for calibration and installation are added to the cost basis - per the statutory cost definition.

For a tax audit, the result is a complete decision trail: why capitalised, how classified, which cost components, which useful life with which legal basis. This documentation is not produced after the fact for the auditor. It arises as a natural by-product of every single capitalisation decision.

That does not reduce the CFO’s responsibility for the accuracy of the balance sheet. It gives them the foundation to exercise that responsibility on an informed basis - instead of trusting that nobody in accounts payable made an error.

Micro-Decision Table

Who decides in this agent?

9 decision steps, split by decider

67%(6/9)
Rules Engine
deterministic
11%(1/9)
AI Agent
model-based with confidence
22%(2/9)
Human
explicitly assigned
Human
Rules Engine
AI Agent
Each row is a decision. Expand to see the decision record and whether it can be challenged.
Identify asset What was acquired and is it a standalone asset? AI Agent Auditor

LLM interpretation of the service description from the invoice

Decision Record

Model version and confidence score
Input data and classification result
Decision rationale (explainability)
Audit trail with full traceability

Challengeable: Yes - fully documented, reviewable by humans, objection via formal process.

Challengeable by: Auditor

Check capitalisation requirement Is the net value above EUR 800 and subject to capitalisation? Rules Engine Auditor

EStG, deterministic threshold check

Decision Record

Rule ID and version number
Input data that triggered the rule
Calculation result and applied formula

Challengeable: Yes - rule application verifiable. Objection possible for incorrect data or wrong rule version.

Challengeable by: Auditor

Determine asset class Which asset class is the asset assigned to? Rules Engine Auditor

Standard cases rule-based (R), borderline cases via LLM suggestion (A)

Decision Record

Rule ID and version number
Input data that triggered the rule
Calculation result and applied formula

Challengeable: Yes - rule application verifiable. Objection possible for incorrect data or wrong rule version.

Challengeable by: Auditor

Determine acquisition costs What are the acquisition or production costs? Rules Engine Auditor

HGB Paragraph 255, invoice amount plus allocable ancillary costs

Decision Record

Rule ID and version number
Input data that triggered the rule
Calculation result and applied formula

Challengeable: Yes - rule application verifiable. Objection possible for incorrect data or wrong rule version.

Challengeable by: Auditor

Assign ancillary acquisition costs Which ancillary costs belong to the asset? Human Auditor

Standard (transport, installation) rule-based (R), complex allocation human (H)

Decision Record

Decider ID and role
Decision rationale
Timestamp and context

Challengeable: Yes - via manager, works council, or formal objection process.

Challengeable by: Auditor

Set useful life What is the standard useful life? Human Auditor

BMF depreciation table (R) for standard assets, special machinery human (H)

Decision Record

Decider ID and role
Decision rationale
Timestamp and context

Challengeable: Yes - via manager, works council, or formal objection process.

Challengeable by: Auditor

Assign inventory number Which inventory number does the asset receive? Rules Engine

Sequential numbering by asset class

Decision Record

Rule ID and version number
Input data that triggered the rule
Calculation result and applied formula

Challengeable: Yes - rule application verifiable. Objection possible for incorrect data or wrong rule version.

Record in asset register Are all mandatory fields complete and consistent? Rules Engine Auditor

Structured recording with completeness check

Decision Record

Rule ID and version number
Input data that triggered the rule
Calculation result and applied formula

Challengeable: Yes - rule application verifiable. Objection possible for incorrect data or wrong rule version.

Challengeable by: Auditor

Assign cost centre and location Which cost centre and location is it assigned to? Rules Engine

Master data from purchase order or organisational assignment

Decision Record

Rule ID and version number
Input data that triggered the rule
Calculation result and applied formula

Challengeable: Yes - rule application verifiable. Objection possible for incorrect data or wrong rule version.

Decision Record and Right to Challenge

Every decision this agent makes or prepares is documented in a complete decision record. Affected parties (employees, suppliers, auditors) can review, understand, and challenge every individual decision.

Which rule in which version was applied?
What data was the decision based on?
Who (human, rules engine, or AI) decided - and why?
How can the affected person file an objection?
How the Decision Layer enforces this architecturally →

Does this agent fit your process?

We analyse your specific finance process and show how this agent fits into your system landscape. 30 minutes, no preparation needed.

Analyse your process

Governance Notes

GoBD-compliant §203 StGB-compliant

Mixed decision distribution (1H / 6R / 1A / 1 R/A). Human decision for special machinery useful life and complex ancillary cost allocations. HGB Paragraph 255 (acquisition cost determination), EStG Paragraph 7 (useful life) and EStG Paragraph 6 Abs. 2 (low-value asset threshold) as direct legal bases.

GoBD-compliant: every asset addition is archived with document, valuation basis and classification. During tax audits, the capitalisation decision is a frequent point of review. Paragraph 203 StGB relevant: asset master data can reveal investments and business models.

§203 StGB-relevant data is encrypted end-to-end and never passed to AI models in plain text.

Process Documentation Contribution

Per asset addition documented: source document (invoice), identified asset, capitalisation decision, asset class with rationale, acquisition cost determination (individual items), useful life with rule source, inventory number, cost centre. For human decisions (special machinery useful life), the clerk's rationale is logged.

Assessment

Agent Readiness 78-85%
Governance Complexity 26-33%
Economic Impact 64-71%
Lighthouse Effect 21-28%
Implementation Complexity 31-38%
Transaction Volume Weekly

Prerequisites

  • Fixed asset accounting with API access (SAP FI-AA, DATEV or equivalent)
  • Current BMF depreciation table (digitally stored)
  • Invoice Capture Agent for pre-classified documents
  • Cost centre structure and location master data

Infrastructure Contribution

The Asset Capitalisation Agent delivers the master data for the Depreciation Agent and Inventory Agent. The asset identification from invoices is a reusable pattern for all agents that must interpret invoice line items. The asset class assignment (rule-based plus LLM support for borderline cases) establishes the hybrid R/A decision pattern common in many Finance agents. Builds Decision Logging and Audit Trail used by the Decision Layer for traceability and challengeability of every decision.

What this assessment contains: 9 slides for your leadership team

Personalised with your numbers. Generated in 2 minutes directly in your browser. No upload, no login.

  1. 1

    Title slide - Process name, decision points, automation potential

  2. 2

    Executive summary - FTE freed, cost per transaction before/after, break-even date, cost of waiting

  3. 3

    Current state - Transaction volume, error costs, growth scenario with FTE comparison

  4. 4

    Solution architecture - Human - rules engine - AI agent with specific decision points

  5. 5

    Governance - EU AI Act, GoBD/statutory, audit trail - with traffic light status

  6. 6

    Risk analysis - 5 risks with likelihood, impact and mitigation

  7. 7

    Roadmap - 3-phase plan with concrete calendar dates and Go/No-Go

  8. 8

    Business case - 3-scenario comparison (do nothing/hire/automate) plus 3×3 sensitivity matrix

  9. 9

    Discussion proposal - Concrete next steps with timeline and responsibilities

Includes: 3-scenario comparison

Do nothing vs. new hire vs. automation - with your salary level, your error rate and your growth plan. The one slide your CFO wants to see first.

Show calculation methodology

Hourly rate: Annual salary (your input) × 1.3 employer burden ÷ 1,720 annual work hours

Savings: Transactions × 12 × automation rate × minutes/transaction × hourly rate × economic factor

Quality ROI: Error reduction × transactions × 12 × EUR 260/error (APQC Open Standards Benchmarking)

FTE: Saved hours ÷ 1,720 annual work hours

Break-Even: Benchmark investment ÷ monthly combined savings (efficiency + quality)

New hire: Annual salary × 1.3 + EUR 12,000 recruiting per FTE

All data stays in your browser. Nothing is transmitted to any server.

Asset Capitalisation Agent

Initial assessment for your leadership team

A thorough initial assessment in 2 minutes - with your numbers, your risk profile and industry benchmarks. No vendor logo, no sales pitch.

30K120K
1%15%

All data stays in your browser. Nothing is transmitted.

Frequently Asked Questions

How does the agent recognise whether an invoice contains an asset or an expense?

The LLM interprets the service description and the threshold check (above EUR 800 net) decides on capitalisation or immediate expense. For low-value assets between EUR 250 and EUR 800, immediate write-off is automatically applied.

What happens with consolidated invoices containing multiple assets?

The agent recognises individual line items and checks each separately for capitalisation. Dependent components are assigned to the main asset. For ambiguity, the item is escalated for human decision.

How are self-produced assets handled?

For in-house production, the agent determines production costs per HGB Paragraph 255 Abs. 2 - including material, manufacturing and allocable overheads. The allocation of complex overhead portions requires human decision.

What Happens Next?

1

30 minutes

Initial call

We analyse your process and identify the optimal starting point.

2

1 week

Discover

Mapping your decision logic. Rule sets documented, Decision Layer designed.

3

3-4 weeks

Build

Production agent in your infrastructure. Governance, audit trail, cert-ready from day 1.

4

12-18 months

Self-sufficient

Full access to source code, prompts and rule versions. No vendor lock-in.

Implement This Agent?

We assess your finance process landscape and show how this agent fits your infrastructure.