Skip to content
W K
GoBD-compliant §203 StGB-compliant Q4

Annual Statement Preparation Agent

Prepare annual financial statements - orchestrate checklist, draft notes, answer auditor queries.

Orchestrates the annual close checklist, consolidates reconciliations, reviews valuation options, drafts notes and management report.

Analyse your process
Airbus Volkswagen Shell Renault Evonik Vattenfall Philips KPMG

Orchestrate closing checklist, LLM draft for notes and report, valuation options with humans

The agent orchestrates the entire annual-statement checklist via a rule engine, drafts notes to the financial statements and the management report via LLM from finance data, and returns valuation options, subsequent-event review and the balance-sheet certification explicitly to executive management.

Outcome: Closing throughput shortened by 30 to 40 percent, 15 documented decision steps per close, and auditor response times accelerated by factor 2 per APQC benchmark.

47% Rules Engine
20% AI Agent
33% Human

Behind each of the 15 steps sits a clear assignment to rule engine, LLM draft or human judgement:

Four-week close, 80 percent routine, 20 percent expert work

No annual financial statement fails because of missing expertise. It fails because of orchestration: reconciliations that never reach completion, queries about the asset register that sit in an inbox for three days, notes disclosures that materialise at the last minute. The finance team knows every single step - and still loses weeks coordinating between them.

The closing weeks consume the entire team on routine work

According to APQC Open Standards Benchmarking, the cross-industry median for the monthly close sits in the higher single-digit days range - most finance teams need longer than five working days for the monthly close alone. The annual statement multiplies that effort: the asset register, provisions schedule, equity schedule, cash flow statement, tax reconciliation, notes, management report, auditor documentation and regulatory filing all come on top.

A concrete scenario illustrates the problem. A mid-sized manufacturer with EUR 200 million (USD 220 million) in revenue closes the fiscal year. The Head of Finance coordinates a team of six over four weeks. Three of them spend most of their time building schedules - arithmetic derivations from data that already exists in the system. One person assembles documents for the statutory auditor. Valuation decisions - the actual expert work - fill perhaps 20 percent of total closing time.

Meanwhile, late filings carry penalties. In Germany, the Federal Office of Justice imposes a minimum of EUR 2,500 (USD 2,750) per offence, rising to EUR 25,000 (USD 27,500) in repeat cases. For capital-market-oriented companies, the reputational risk can outweigh the fine itself. (UK: Companies House late filing penalties follow a similar tiered structure.)

Fifteen decision steps separate routine from judgement

The Decision Layer decomposes the entire annual statement process into fifteen individual decision steps - and assigns each to a clear tier. The result is a map showing where automation adds value and where human judgement remains essential.

Tier 1 (rule engine) covers nine steps: the close checklist, consolidation of reconciliations, the tax reconciliation for standard items, deferred tax calculations based on temporary differences, the filing preparation, and the four schedules (provisions, assets, equity, cash flow). All of these follow defined rules and existing data.

Tier 2 (AI draft with human approval) covers three steps: the notes disclosures draft, the management report draft and the preparation of auditor documentation based on the PBC (Prepared by Client) list. The agent produces structured drafts - content responsibility stays with the specialist team.

Three steps remain exclusively with the human: valuation options (an accounting policy decision), the subsequent events review (a materiality assessment) and the balance sheet oath (personal attestation by management). No automation replaces the judgement these steps require.

The agent orchestrates - the human decides

What changes in practice? The agent takes over sequencing. It checks whether all reconciliations are complete before triggering the next step. It builds the four schedules directly from the outputs of upstream agents. It prepares notes disclosures as a structured draft the team revises rather than writes from scratch. It assembles the auditor package according to the PBC list.

The team shifts its time: away from coordination, toward valuation decisions. The question is no longer “Is the asset register finished?” but “Which valuation method do we apply to pension provisions?” - the question a CFO is actually paid to answer.

For the statutory auditor, a side effect emerges: every decision step is documented, every data source traceable. The audit does not become less rigorous - but it starts from a complete, structured foundation.

Every upstream agent improves closing quality

The Annual Statement Preparation Agent is not an isolated tool. It is the integration point of the entire finance infrastructure. The asset register draws on the Depreciation Agent. The provisions schedule relies on the Provisions Agent. Account reconciliations come from the Reconciliation Agent, tax data from the VAT Agent, write-downs from the Receivables Management Agent.

The implication: organisations that run the upstream agents cleanly have better data at annual close - automatically. Those that define the annual statement as the goal build the entire agent infrastructure backwards from that point. The annual statement is not the beginning - it is the proof that the infrastructure works.

Micro-Decision Table

Who decides in this agent?

15 decision steps, split by decider

47%(7/15)
Rules Engine
deterministic
20%(3/15)
AI Agent
model-based with confidence
33%(5/15)
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.
Orchestrate close checklist Which steps must be completed in which order? Rules Engine

Configured close checklist with dependencies

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.

Consolidate reconciliations Are all accounts finally reconciled? Rules Engine Auditor

Status check of all reconciliation reports

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

Review valuation options Which valuation methods are applied for HGB and IFRS? Human Auditor

Accounting policy decision with strategic impact

Decision Record

Decider ID and role
Decision rationale
Timestamp and context

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

Challengeable by: Auditor

Draft notes disclosures How are mandatory disclosures formulated in the notes? AI Agent Auditor

LLM creates draft from financial data and prior-year notes

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

Draft management report How is the business situation presented in the management report? AI Agent Auditor

LLM creates draft from KPIs and management commentary

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

Tax reconciliation How is the reconciliation from commercial to tax accounts made? Human Auditor

Standard items = R, deviations requiring judgement = 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

Calculate deferred taxes Are deferred tax assets recognised? Human Auditor

Calculate temporary differences = R, activation decision = 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

Subsequent events review Are there value-clarifying or value-creating events after the balance sheet date? Human Auditor

Assessment of materiality and attribution

Decision Record

Decider ID and role
Decision rationale
Timestamp and context

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

Challengeable by: Auditor

Prepare auditor queries How are typical auditor queries answered? AI Agent Auditor

LLM creates draft answers from financial data

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

Balance sheet oath Is the management representation letter signed? Human Auditor

Personal attestation by management

Decision Record

Decider ID and role
Decision rationale
Timestamp and context

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

Challengeable by: Auditor

Prepare filing Is the annual statement filed with the Federal Gazette? Rules Engine

Format specification of the Federal Gazette

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.

Create provisions schedule How did provisions develop during the fiscal year? Rules Engine Auditor

Arithmetic compilation

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

Create asset register schedule How did fixed assets develop during the fiscal year? Rules Engine Auditor

Arithmetic compilation

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

Create equity schedule How did equity develop during the fiscal year? Rules Engine Auditor

Arithmetic compilation

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

Create cash flow statement How is cash flow calculated by the indirect method? Rules Engine Auditor

Indirect method = R, plausibility check = 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

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

The annual statement is subject to HGB Paragraph 242 ff. (obligation to prepare) and must comply with generally accepted accounting principles. Corporations must file per HGB Paragraph 325 with the Federal Gazette. The tax reconciliation follows EStG and KStG.

Deferred taxes are treated per HGB Paragraph 274 (recognition option for deferred tax assets) or IAS 12 (recognition obligation). The subsequent events review follows HGB Paragraph 252 Abs. 1 Nr. 4. The balance sheet oath per Paragraph 289 Abs. 1 Satz 5 HGB is a personal declaration by management.

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

Process Documentation Contribution

The Annual Statement Preparation Agent documents for the GoBD procedural documentation: the complete process of annual statement preparation, which valuation options were exercised, how deferred taxes were calculated, which subsequent events were identified and how they were treated.

Assessment

Agent Readiness 42-49%
Governance Complexity 51-58%
Economic Impact 68-75%
Lighthouse Effect 46-53%
Implementation Complexity 58-65%
Transaction Volume Yearly

Prerequisites

  • Completed monthly close for all 12 months
  • Complete reconciliation reports for all accounts
  • Prior-year annual statement as comparison basis
  • Auditor programme (optional, accelerates preparation)

Infrastructure Contribution

The Annual Statement Preparation Agent is the convergence point of all Finance agents. It uses the reconciliation reports from the Reconciliation Agent, provisions from the Provisions Agent, the asset register from the Depreciation Agent and close orchestration from the Close Checklist Agent. The LLM draft for notes and management report becomes the template for subsequent years.

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.

Annual Statement Preparation 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

Can the agent fully automate the annual statement?

No. The agent automates the mechanical steps and prepares drafts. Valuation options, deferred taxes, subsequent events review and the balance sheet oath require human judgement. For a typical mid-sized company, manual effort is reduced by 40-60%.

How does the agent support the financial audit?

The agent prepares answers to typical auditor queries, compiles all reconciliation reports and provides the decision record for every posting-relevant decision. This significantly shortens the financial audit because the auditor receives all evidence immediately.

Does the agent support both HGB and IFRS statements?

Yes. The agent supports parallel accounting standards. Valuation differences between HGB and IFRS are documented and shown in the reconciliation. Notes disclosures are formulated differently per standard.

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.