Claude Cowork builds weekly overdue-invoice drafts
Michel reports a Claude Cowork-built weekly workflow that matches overdue accounts with a contact Sheet, prepares personalised invoice reminders as drafts and sends him a customer-and-total report.
Suggested prompt
Suggested prompt written by Agents In Use, not the source author. Replace bracketed inputs and review tool permissions first.
Help build a Monday-night workflow that checks overdue accounts, matches them to my customer-contact Sheet and fills my reminder template with invoice ID, balance and due date. Save drafts and a coverage-and-total report for my review; I will send approved reminders.
Reach people exploring AI agents
See placements →What happened
Michel reports a Claude Cowork-built weekly workflow that matches overdue accounts with a contact Sheet, prepares personalised invoice reminders as drafts and sends him a customer-and-total report.
Reported by MichelHow the agent did it
- Prepare the customer-to-contact Google Sheet and reminder template.
- Use Cowork to construct the scheduled overdue-accounts workflow.
- Match overdue customers and populate invoice ID, amount and due date.
Where a person stepped in
Michel reviews drafts and presses Send. Reported time savings are personal, and the source does not prove payment recovery or an ambiguity-handling method.
Michel has a limited back-office system and regularly spends Monday evenings preparing overdue-invoice reminders. Claude Cowork helps him create an agent workflow that logs into the system, checks late accounts receivable and compares customer names with a Google Sheet he supplies.
The workflow uses his email template and adds the contact name, invoice identifier, amount and due date. It places finished messages in the Drafts folder rather than sending them to customers. A separate report reaches Michel with the customers included, each balance and the total, giving him a second view of what the draft batch covers.
He reviews the messages, confirms the contents and presses Send himself. The account reports about an hour saved each Monday, without an independently measured time study or evidence of debt recovered.
Claude Cowork is specifically credited with constructing the agent workflow. The post does not identify its running harness, so this entry does not claim Cowork itself performs every scheduled login. The useful operational detail is separating reminder preparation from customer delivery and including both invoice-level identifiers and an aggregate report. Name matching still needs review: the source does not describe a demonstrated fallback for ambiguous names, missing contact rows or disputed invoices. Those limits matter before adapting the process to a different accounts system.
Prompt
Suggested prompt (written by Agents In Use, not the author's)
Source: Blog · Michel
Frequently asked questions
What makes the reminders specific?
The supplied template is populated with contact name, invoice identifier, amount and due date.
Who sends the customer messages?
Michel reviews the Drafts folder, confirms the contents and presses Send himself.
Does the report prove debt was recovered?
No. It describes prepared reminders and their balances and totals, not a measured collection outcome.
More details
Remaining steps
- Place customer reminders in Drafts and deliver the coverage-and-total report.
- Review each draft and personally send approved reminders.
About this account
Cowork helps build the scheduled agent workflow, which reads late accounts, matches contacts, prepares reminder drafts and reports its coverage; the runtime harness is unnamed.
Firsthand account by Michel; original publication date verified against Primary post JSON-LDdatePublished2026-10-13T04:58:41.773Z and time datetime2026-10-13T04:58:41.773984Z. Outcomes are the author’s reports, not independently reproduced. Suggested prompt written by Agents In Use; not the author’s original prompt.
Credit: Michel
Source published 13 Oct 2026
Source checked 14 Oct 2026
Tools and topics
Claude Cowork · Accounts-receivable system · Google Sheets · Email Drafts · Coverage report
Overdue reminder drafts · Customer-contact matching · Invoice batch totals
Editorial completeness: 100/100. Measures documented fields, not agent performance or verified outcomes.