Cash Application — Koko Knows

What changed

Bank files parse by rule, and a language model now extracts the invoice references from the messy email and PDF remittances that used to need manual keying. Matching against open invoices then applies most payments straight through, and only short pays and true unmatched items reach a person.

How it's built

LLM workflow (rung 4 of 6). A coded control path: chain, router, parallel fan-out, evaluator-optimizer.

Not an agent: the sequence of steps is fixed in advance, so the whole path can be tested end to end.

A language model handles step 2. Rules: steps 1, 4 and 5 · Predictive ML: step 3.

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: The path is testable end to end. Each step is independently evaluable.

What the certification exams call this

Anthropic (Claude Certified Architect – Foundations (CCAR-F))Workflow: models and tools on predefined code paths (chaining, routing, parallelization, evaluator-optimizer). source
OpenAI (OpenAI agent guidance (Agents SDK, Codex, OpenAI Academy))Orchestrating via code, which the Agents SDK calls more deterministic and predictable. source
Google Cloud (Generative AI Leader)Workflow agents (sequential, parallel, loop) that run steps in a fixed order. source
Microsoft (AI-901 (Azure AI Fundamentals) and AB-100 (Agentic AI Business Solutions Architect))Sequential and concurrent orchestration patterns, with steps in a set order. source

How Koko classifies AI · the full ladder is taught in AI & Agentic Delivery Foundations.

Modeled value on KokoAI

Synthetic KokoAI figures — a demo fixture, not audited or a guarantee

Straight-through cash application

60 → 92 %

higher is better

Recurring impact

$2M/yr

run-rate benefit

Cash freed

$41M

one-time working-capital release

Payback

3 months

≈ 4,000 hrs/yr reclaimed

Auto-application 60%→92% clears ≈3 days of unapplied cash (≈$41M) into the ledger sooner; ≈$2M/yr yield on cash recognized earlier plus faster, cleaner reconciliation.

Run this in Claude on KokoAI's data

Opens your own Claude with the prompt and reference data pre-loaded as text — no upload, no setup. Tailor it live to your own numbers.

Try it

Illustrative demo

Synthetic data

Worked example of "Cash Application" on synthetic data — edit the inputs to tailor it. Edit the inputs to tailor the output — it's baked synthetic data, so nothing leaves your browser until you run it live.

Company / entity

Synthetic demo company (a ~$5B high-growth US AI SaaS) — swap in any name to tailor the output.

Period

Focus / scope

What this run should concentrate on.

Run live in Claude

Workflow

1.RulesCapture payment data from bank files, parsed by rule, and remittances from email and portals.
2.Generative AIExtract invoice references and amounts from unstructured remittances, and normalize them.
3.Predictive MLMatch payments to open invoices and score each match; apply those above the approved confidence threshold.
4.RulesPost matched cash automatically; queue short pays and unmatched items for review.
5.RulesHand off short pays and deductions to the dispute workflow.

Prompt recipe

Parse the attached remittance and match each payment to the open invoices in the AR ledger. Apply matches with confidence ≥ [threshold]; for partial or unmatched payments, show candidate invoices and the gap, and label likely short-pays vs. on-account.

Try in Ask KokoAI

Controls required

Confidence threshold for straight-through posting approved
Unmatched/short-pay cash held for human review
Bank reconciliation confirms applied cash
Audit trail of every match and posting