Touchless AP & Invoice Matching
2/3-way match supplier invoices and auto-route only the exceptions.
A document-AI model reads supplier invoices, deterministic logic performs the 2- and 3-way match against PO and receipt within approved tolerances, and a trained model flags duplicates and anomalies. Only true exceptions reach a person. The core flow needs no generative model.
Predictive ML (rung 1 of 6). Score, forecast, classify, cluster or detect anomalies from your own history.
Not an agent: the sequence of steps is fixed in advance, so the whole path can be tested end to end.
No language model in the path: every step below runs by rule, by a trained model, or by a person.
- Autonomy: T2 · Act on exception
- Acts unattended on the confident majority; routes the rest to a human. Human role: reviews only the routed cases.
- How it is governed
- Model risk management: validation, monitoring, bias testing, periodic revalidation.
- Anthropic (Claude Certified Architect – Foundations (CCAR-F))Outside the Claude exams. A trained score is data or a tool result the model reads.
- OpenAI (OpenAI agent guidance (Agents SDK, Codex, OpenAI Academy))Outside the agent guide. A trained model is a data source or tool.
- Google Cloud (Generative AI Leader)Machine learning: supervised, unsupervised and reinforcement learning (exam §1.1). source
- Microsoft (AI-901 (Azure AI Fundamentals) and AB-100 (Agentic AI Business Solutions Architect))Custom AI models: AB-100 asks when one should be created. source
How Koko classifies AI · the full ladder is taught in AI & Agentic Delivery Foundations.
Synthetic KokoAI figures — a demo fixture, not audited or a guarantee
Straight-through 3-way match 45%→86% across ≈240k invoices/yr cuts processing cost ≈$3.2M/yr; duplicate + price-variance catches (e.g. the $3.12M CS-99120 duplicate) avoid ≈$1.3M/yr — ≈$4.5M/yr recurring.
Opens your own Claude with the prompt and KokoAI's supplier invoice batch pre-loaded as text — no upload, no setup. Tailor it live to your own numbers.
Try it
Illustrative demoA batch of KokoAI supplier invoices runs through 3-way match: clean ones auto-approve, exceptions (price variance, missing receipt, duplicate) route to a human. Edit the inputs to tailor the output — it's baked synthetic data, so nothing leaves your browser until you run it live.
Price/quantity tolerance for straight-through approval.
Supplier Invoice Batch (synthetic sample) — grab the one-pager PDF (the ROI, the data, and the prompt on one sheet) to share or keep, or the raw CSV / Markdown to drag into Claude or Microsoft Copilot. The prompt hand-off above already includes the same data as text.
Synthetic KokoAI demo data — not audited
Drop a PDF, Excel, Word, CSV, or text file (e.g. a 10-Q, an AR aging, an invoice register) and we'll run this use case on it. Your file is read in your browser and never uploaded — only the extracted text goes into your own Claude or Copilot.
Read in-browser · not uploaded · 10-Q/10-K trimmed to financials + MD&A
Workflow
- 1.Predictive MLExtract invoice header and line data with a document-AI model.
- 2.RulesPerform the 2-way (invoice↔PO) and 3-way (↔goods receipt) match.
- 3.RulesAuto-approve clean matches within tolerance; route exceptions.
- 4.RulesApply payment terms and schedule payment for approved invoices.
- 5.Predictive MLDetect duplicate or anomalous invoices and hold them for review.
Prompt recipe
Extract the line items from the attached invoice and match them to the referenced PO and goods receipt. Approve within [tolerance]; for any mismatch (price, quantity, missing receipt) explain the exception and route it. Flag possible duplicates against recent invoices.
- Match tolerances and auto-approval limits approved
- Duplicate-payment and fraud checks before payment
- Segregation of duties between approval and payment run
- Exception and override audit trail