Building a resilient n8n automation workflow — need advice on error handling

I’m currently building an automation system in n8n that connects multiple APIs to handle lead processing and data enrichment automatically. The workflow takes incoming leads, cleans and formats the data, enriches it using external APIs, and then routes it to different destinations depending on certain conditions.

Right now, I’m running into a challenge with handling retries and error states properly when one of the API nodes fails mid-flow. I want the workflow to be resilient without duplicating data or breaking the sequence.

What’s the best practice you use in n8n for handling API failures and retry logic inside complex multi-step workflows?

4 comments

Building a resilient n8n automation workflow — need advice on error handling