I’m trying to set up an automated workflow that can delete related entries from my PostgreSQL database whenever I remove a record from Airtable. I’m using Zapier to implement this, but I’ve encountered a problem.
The triggers available for Airtable in Zapier only include options for creating new records or updating existing ones. Unfortunately, I can’t find any built-in trigger for when records are deleted.
Has anyone figured out a workaround for this? Perhaps there is a different method or a clever solution utilizing the triggers that are available? It’s essential for me that my PostgreSQL database stays consistent with my Airtable records, so any elimination in Airtable should reflect in the corresponding PostgreSQL table.
I would greatly appreciate any advice or tips. Thank you!
Skip hard deletes in Airtable - use soft deletes instead. I added a status field with ‘active’ and ‘archived’ values. When you need to delete something, just flip it to archived. Set up a Zapier trigger for record updates, filter for status changes to archived, then delete the PostgreSQL entry. You keep the audit trail in Airtable but stay synced. Downside? Archived records pile up, but you can clean them out manually if storage gets tight.
I’ve dealt with this same issue syncing Airtable and PostgreSQL. Here’s what worked for me: I set up a daily automation that compares records between both systems. I keep a timestamp field in Airtable that updates whenever someone changes a record, and my PostgreSQL table stores the matching record IDs. The automation checks both databases, finds any PostgreSQL entries that don’t exist in Airtable anymore, and deletes them. It’s not real-time, but this batch method keeps everything consistent. If you need faster syncing, you could use Airtable’s API with a webhook service - just know that means building something outside Zapier.
i get you’re frustraited, miar! i had a similar issue and what worked for me was using a “deleted” checkbox in airtable. checking it lets zapier know to delete the related entry in postgres. not ideal, but it does the job!