Route customer-data conflicts into an Airtable review queue
At a glance
An Indie Hackers tutorial connects standardized intake to AI checks and alerts for inconsistent customer records. This workflow is listed under Customer Operations, Data Quality, Automation.
SourceBuilding a tool for these workflows? See weekly rates.
What they did
1) Connect Jotform, Make, Airtable, OpenAI and Gmail. In Airtable, create a Customers table with identity, plan, payment status, notes, conflict, missing-field, review and update-time fields. Make email, plan and payment status required on a matching Jotform. 2) Send submissions to a Make custom webhook. Add an OpenAI step that returns JSON with trimmed fields, lowercase email and email as the fallback customer ID. 3) Search Airtable by email; route matches to Update Record and new customers to Create Record. 4) Pass the resulting record to another OpenAI step checking missing values, duplicate indicators and plan/payment contradictions. Return an issues list, suggested fixes and a flag. 5) Save those results, set Needs review when flagged and send a Gmail alert containing the issues and record link. Submit a test intake and inspect its route and alert.
Worth knowing
Stripe payment-event syncing is optional; the tutorial also adds a scheduled weekly summary of records awaiting review.