--- name: al-662-stop-chasing-every-late-invoice-the-same-way-put-ai-colle description: "PYMNTS (Oct 20, 2026) reports Nestle is adopting SAP's Joule Accounts Receivable Assistant as it builds a 140-country SAP S/4HANA core. Nestle's Luca Dell'Orletta said the assistant can prioritize collections by risk and business context and recommend next-best actions. SAP's product page describes specialized agents under an orchestration agent: Collection Account Preparation, Collections Email Outreach, Dispute Creation, Dispute Resolution, plus related AR agents for dunning insights, account analysis and clearing. SAP states benefits around faster cash collection and dispute resolution but has published no Nestle performance figures or a firm general-availability date for the assistant. Treat vendor benefit claims as unverified until your own pilot measures DSO and dispute cycle time. Use when the user wants help with multi-agent accounts receivable collections (finance) for a small business." --- # Stop chasing every late invoice the same way: put AI collections agents on prioritization, call prep, follow-up email, dispute creation and dispute resolution so collectors work the highest-risk accounts first PYMNTS (Oct 20, 2026) reports Nestle is adopting SAP's Joule Accounts Receivable Assistant as it builds a 140-country SAP S/4HANA core. Nestle's Luca Dell'Orletta said the assistant can prioritize collections by risk and business context and recommend next-best actions. SAP's product page describes specialized agents under an orchestration agent: Collection Account Preparation, Collections Email Outreach, Dispute Creation, Dispute Resolution, plus related AR agents for dunning insights, account analysis and clearing. SAP states benefits around faster cash collection and dispute resolution but has published no Nestle performance figures or a firm general-availability date for the assistant. Treat vendor benefit claims as unverified until your own pilot measures DSO and dispute cycle time. ## Inputs to collect first Ask the user for anything below that they haven't already given you. Don't guess missing facts. - Number of open overdue invoices, total overdue $, top reasons… - Days overdue, amount, customer segment, dispute flag, promise-to-pay ## Instructions Do the task below with the user's inputs in place of each [bracket]. Be my collections-desk planner for overdue invoices. Here's our AR snapshot: [number of open overdue invoices, total overdue $, top reasons disputes happen, tools: ERP/CRM/email]. 1) Propose 4 specialized AI agents (account prep, email outreach, dispute create, dispute resolve) and what each may read vs what a human must approve before send or case create. 2) Design a daily worklist score using only fields we have ([days overdue, amount, customer segment, dispute flag, promise-to-pay])--show the ranking formula in plain words, no invented weights pretending to be Nestle's. 3) Draft one call-prep brief template and one follow-up email template with merge fields, plus a rule that no email sends without a person clicking send. 4) For dispute create: checklist of documents to extract and how to flag duplicate invoices. 5) Give a 30-day pilot plan: which segment to start with, what to measure (DSO, disputes closed, emails sent vs replied), and a kill switch if error rate rises. Never move money or change credit holds yourself. ## Safety checks (share these with the user) - Check every figure, price and total against your own records before anything is sent, posted or booked. A person approves refunds, discounts, payments and price changes. - Remove names, contact details, account numbers and other personal data the task doesn't need before you paste, and check your AI tool's data and training settings. - Read and edit the draft before it reaches a customer or goes public. If you later automate this prompt, start with "draft and notify me", not "send automatically". - Check names, dates, numbers and any claim against your own source material before you use the answer. AI can misread or invent details. - Say clearly which parts need a person to review or approve before anything is sent, paid, posted or filed. ## About this skill - Works with: SAP Joule Accounts Receivable Assistant (or similar ERP AI), ChatGPT, Claude, your ERP and email - Based on this real-world example: https://www.pymnts.com/news/artificial-intelligence/2026/nestle-hands-its-overdue-invoices-to-saps-ai-agents/ - Free prompt from AgentPlaybook Real Estate: https://agentic-library.netlify.app/prompt/al-662-stop-chasing-every-late-invoice-the-same-way-put-ai-collecti/ - To install in an agent that supports Agent Skills, save this file as SKILL.md in a folder with the same name as `name` above (for example in your skills directory). Install steps for Claude.ai, Claude Code, Cursor and Codex: https://agentic-library.netlify.app/skills/