The AI Reconciliation Layer for enterprise finance.
Flexibly adapts to your ecosystem, so every line item is reconciled in real time.
ISO 27001
EU hosted · Frankfurt
GDPR
4–6 weeks demo to live
Cash Application
·
RECON
Live
ERP
·
Open AR
5 items
·
€96,522.50
INV-29481
€48,210.00
INV-29482
€12,980.00
INV-29483
€7,422.50
INV-29485
€24,100.00
INV-29486
€3,810.00
Bank
·
Incoming
1 payment
·
€95,662.50
Matched
€48,210.00
Matched
€12,980.00
Matched
€7,422.50
SHORT-PAY
€23,240.00
Matched
€3,810.00
%+
Automation rate
Confidence score on every match.
%
Line-item coverage
No sampling.
1–3
%
Leakage identified
Measured on your data.
4–6
weeks
Demo to live
No big project.
Configured to your data and systems
The architecture behind THE LAyer
Flexible AI Reconciliation
Donnerstag.ai is the AI Reconciliation Layer that sits between the systems that record financial transactions and the finance teams that depend on those records being true.
We use AI to flexibly adapt to every customer's systems and processes so data ingestion, matching, dashboards and the audit trail become fully customized and automated in 4-6 weeks. Not in months.
your Data and Systems
Automated data source ingestion
All of your data sources ingested into one layer. Structured or unstructured. ERP, PDF, EDI, bank, expense, TMS.
Flexible
Customized matching
Customized deterministic + AI-authored matching rules at line-item level. Confidence score on every match.
Customizable
Real-time dashboards
Every transaction reconciled on the day it posts. Customizable dashboards report in real-time.
Audit ready
Automatic audit trail
Every decision logged immutably. Rule, model, timestamp, user, confidence, rationale.
Reconciliation is inherently customer-specific
Configured to your data and systems.
No two businesses have the same data sources, the same matching rules, or the same exception patterns. We flexibly adapt to every customer-specific case. A solution that fits your ecosystem.
01
Read from your systems of record
ERP, bank, EDI, PDF, TMS, corporate card, T&E feeds. Native connectors, one data model. No middleware, no schema forcing.
02
The engine writes rules against your data
The engine customizes deterministic rules to your specific patterns. Where the patterns are non-obvious, the engine drafts the match rule and your team reviews and approves it. The output is transparent, auditable, yours to extend.
03
Write back, additive and removable
Only team-approved postings into your GL and sub-ledgers the way your auditor expects. Read-from / write-to. Nothing is rip-and-replace, everything is additive.
AI customizes. The matching runs on rails.
Your matching rules. Customized by AI.
AI writes code, not data. The Matching Engine customizes deterministic rules to your specific patterns. When it encounters a non-obvious pattern in your data, it drafts a rule. Your team reviews, adjusts, and approves the rules.
Human-on-the-Loop by design, because auditors read rules, not model outputs.
Customized deterministic rules
Exact, tolerance, and multi-leg matches. Transparent logic your auditor can read and your team can extend. Rules first, always.
AI-drafted rules
Where the pattern in your data is non-obvious, the engine drafts a rule against it. The team reviews, adjusts, and approves it. The rule is the output, not a black-box decision.
Team review on exceptions
Genuine exceptions stay with your team, where judgement belongs. Every override is attributable: rule, user, timestamp, rationale.
The Layer, delivered through Recons
The RECON Library.
Each RECON gets customized so it covers your reconciliation use case end-to-end and strengthens the Layer's coverage without rebuilding anything underneath.
Start with the RECON you need. Build your Layer over time.
Accounts Receivable
Cash Application RECON
Deduction Handling RECON
Short-Pay RECON
Chargeback RECON
Customer Rebate RECON
Statement Matching RECON
Accounts Payable
Three-Way Match RECON
GR/IR Clearing RECON
Vendor Rebate RECON
Duplicate Invoice RECON
Vendor Dispute RECON
Statement Matching RECON
4 pillars · one architecture · cause and consequence
What you get with the AI Reconciliation Layer
01
Reconciliation Confidence
Full reconciliation confidence
Today's reconciliation processes sit below the achievable optimum. The Reconciliation Layer changes that. Automated Data ingestion, normalization, and 100% line-item coverage in real-time stops being a target and becomes the operating condition. Every match carries a confidence score, every exception is raised on the day it appears.
95%+
Automation rate
100%
Line-item coverage
Hours, not weeks
Exception backlog cleared
02
Revenue & Margin Recovery
Recover what's lost
At the total, offsetting errors cancel and the loss stays invisible. At line-item, the loss becomes visible on the day it occurs. Dispute windows preserved, not discovered late.
1–3%
Revenue recovered
30% → >80%
Dispute success rate
0.5–2%
Margin gain on AP
03
Audit-Ready Close
An automated audit trail
Every matching decision logged immutably: rule, model, timestamp, user, confidence, rationale. GoBD, IFRS, SOX stop being a project and become an operating condition.
100%
Line-item coverage
Up to −50%
Year-end audit time
GoBD · IFRS · SOX
One integrated trail
04
Real-Time Control
A real-time control plane
AR, AP, bank, intercompany: reconciled throughout the month. When the close begins, everything is already reconciled. Month-end becomes a confirmation, not a catch-up. Treasury plans on reliable data.
up to −60%
Month-end close time
1–2 days
IC consolidation
−90%
Manual work
Validate the business case on your data
Quantify on your data.
Real data of one month of your choice. We configure and implement the first RECON, review findings, measure match-rate, money-at-risk, close acceleration potential, and audit exposure on real transactions.