# Mine pharma complaint trends and trigger CAPA alerts: n8n workflow

> Started by a clock, and works across Slack. It hands the judgement calls to Claude. 19 steps, 2 branches: a working afternoon to understand properly.

Source: https://millionrounds.com/workflow/mine-pharma-complaint-trends-and-trigger-capa-alerts (Millionrounds; every figure on the page links to its filing or article)

[Automation library](https://millionrounds.com/workflows) · [Customer support](https://millionrounds.com/workflows?category=Customer+support#library)

# Mine pharma complaint trends and trigger CAPA alerts

Started by a clock, and works across Slack. It hands the judgement calls to Claude. 19 steps, 2 branches: a working afternoon to understand properly.

Medium Calls a model: Claude

[Download free](https://millionrounds.com/login?next=%2Fapp%2Fworkflow%2Fmine-pharma-complaint-trends-and-trigger-capa-alerts)

**19 steps** Medium: a working afternoon

**2** Branches: places the flow splits

**0** Loops or waits

**1** Service it connects to

## The flow

Starts it 2 Model 2 Logic 11 Service 4

The steps in the order they run, from what starts it.

- 1 **Schedule Trigger - Poll Complaint Source** Schedule trigger *Starts it*
- 2 **Manual Trigger - Test Run** Manual trigger *Starts it*
- 3 **Anthropic - Complaint Extraction Model** Lm Chat Anthropic *Model*
- 4 **Set - Config** Set *Logic*
- 5 **AI Agent - Extract Structured Complaint Fields** Agent *Model*
- 6 **HTTP - Fetch New Complaints** Http Request *Service*
- 7 **JS - Parse Extracted Fields** Code *Logic*
- 8 **JS - Normalize Complaint Records** Code *Logic*
- 9 **JS - Cluster by Product & Defect Category** Code *Logic*
- 10 **IF - New Complaints Found?** If *Logic*
- 11 **JS - Compare Clusters Against Rolling Baseline** Code *Logic*
- 12 **JS - Fan Out Complaints for Extraction** Code *Logic*
- 13 **No Action - End Cycle** No Op *Logic*
- 14 **JS - Fan Out Clusters for Routing** Code *Logic*
- 15 **Switch - Route by Statistical Anomaly** Switch *Logic*
- 16 **HTTP - Create CAPA Quality Escalation** Http Request *Service*
- 17 **Merge - Combine Escalation & Routine Paths** Merge *Logic*
- 18 **Slack - Alert Quality Team (Cluster Escalation)** Slack *Service*
- 19 **HTTP - Log Cluster to Trend Store** Http Request *Service*

## How it's built

- **Starts on:** A clock
- **Steps:** 19, with 2 branches
- **Size:** Medium: a working afternoon
- **Connects to:** Slack
- **Model:** Claude

## Questions

What does it need to run?

An n8n account, plus your own Slack account, and a key for Claude. You connect your own accounts after importing it; none are included.

What starts it?

It starts on a clock.

Does it use AI?

Yes. It calls Claude. Model calls are billed by the provider, not by n8n.

How long does it take to set up?

Plan an afternoon. 19 steps is enough that you should read each one before running it.

How do I import it?

Copy the workflow, open a new workflow in n8n and paste. The steps appear on the canvas, ready for your own accounts.

**Before you run it**

Nothing here has been tested by us. It is somebody's working automation, read and described from its own steps. Treat the connections it wants as a list of accounts it will touch, and read the 19 steps before you point it at anything that matters.

## Run it in n8n

- **Create your n8n account** Free to start. n8n is where the workflow runs. [Create an n8n account ↗](https://millionrounds.com/n8n)
- **Download the workflow** Free with a Millionrounds account. [Download free](https://millionrounds.com/login?next=%2Fapp%2Fworkflow%2Fmine-pharma-complaint-trends-and-trigger-capa-alerts)
- **Paste it into n8n** Open a new workflow in n8n and paste. Then connect your own accounts.

Stored with the author's traces removed: canvas notes, instance identifiers, saved credential names and any live document they had left behind.

## Connects to

Each opens the other workflows that use it.

[Slack](https://millionrounds.com/workflows?service=Slack#library)

## Workflows like it

Also customer support, with the same services where possible, closest in size first.

- [**Triage and resolve support tickets with OpenAI** 19 steps · Slack](https://millionrounds.com/workflow/triage-and-resolve-support-tickets-with-openai)
- [**AI sentiment-driven priority classification** 18 steps · calls a model · Google Sheets, Slack](https://millionrounds.com/workflow/ai-sentiment-driven-priority-classification)
- [**Run an AI student helpdesk with OpenAI and Google Sheets** 20 steps · calls a model · Gmail, Google Sheets](https://millionrounds.com/workflow/run-an-ai-student-helpdesk-with-openai-and-google-sheets)
- [**Triage support tickets with OpenAI and HubSpot** 20 steps · Gmail, Google Sheets](https://millionrounds.com/workflow/triage-support-tickets-with-openai-and-hubspot)
- [**Analyze pentest reports and create remediation tickets** 17 steps · calls a model · Slack](https://millionrounds.com/workflow/analyze-pentest-reports-and-create-remediation-tickets)
- [**Analyze support screenshots with UploadToURL and OpenAI** 17 steps · calls a model · Jira, Slack](https://millionrounds.com/workflow/analyze-support-screenshots-with-uploadtourl-and-openai)