P2P baseline
23% invoices failing 3-way match3-way match exception pressure.
A live model of queues, root causes, preventable work, agent intervention, and value leakage.
3-way match exception pressure.
GRIR pressure targeted upstream.
Manual cash application volume.
Unapplied cash baseline.
| Flow | Volume / baseline | Root cause | Classification | Recommended intervention | Value lever |
|---|---|---|---|---|---|
| P2P invoice intake | 12 active exceptions | Missing GR/SES, tolerance variance, duplicate risk | 38% preventable | Policy simulation plus GR prompt | Effort avoided, GRIR reduction |
| P2P duplicate control | $139M annual duplicate risk | Supplier, PO, amount, prior payment match | Judgment/control | Hold, evidence pack, supervisor route | Leakage stopped |
| O2C cash application | 41% of ~34k/month, roughly 14k manual | No remittance, short-pay, deductions | 32% preventable | Structured remittance before cash lands | Cash unlocked, effort avoided |
| O2C deductions | $116M | Inconsistent deduction taxonomy | Standardizable | 15 global categories and owner routing | DSO and query reduction |
| Layer | Purpose | Examples |
|---|---|---|
| Prevent | Stop avoidable exceptions entering the queue. | Tolerance bands, remittance mandate, data-quality gates. |
| Standardize | Normalize variants so agents can reason safely. | Payment terms, deduction taxonomy, master data readiness. |
| Automate | Route clean repeatable work to bounded agents. | Cash match, invoice match, standard evidence assembly. |
| Govern | Keep material decisions under policy and human accountability. | Approval, control, replay, value attestation. |
| Current state | Runtime intervention | Future state |
|---|---|---|
| Manual AP exception queue | Policy simulation plus receiving prompt | Prevented before VIM exception |
| Unidentified cash investigation | Proactive remittance and match proposal | Clean match or governed deduction |
| Email and spreadsheet chase | Case evidence, owner, SLA, value | Prioritized intervention backlog |