Flag repeat customer contacts in Gorgias and alert Slack using Claude Cowork

Prerequisites

Claude Desktop with Cowork enabled
Gorgias account email, API key, and domain
Slack Incoming Webhook URL for your support escalation channel

Overview

Claude Cowork lets you schedule a recurring task that checks for repeat contacts — customers who have opened 3 or more tickets in the past 7 days. Claude fetches recent tickets from the Gorgias API, groups them by customer, identifies repeat contacts, tags their latest ticket, and posts a Slack alert. No code to write or maintain — everything runs from the task description.

This is the fastest way to set up repeat contact detection without any engineering work.

Step 1: Create a "repeat-contact" tag in Gorgias

Before setting up the task, create the tag you'll use to flag tickets:

Go to Settings → Tags in Gorgias
Click Create tag
Name it repeat-contact
Optionally set a color (red or orange works well for attention)

This tag lets you create a Gorgias View that shows all flagged tickets in one place.

Step 2: Set up the Slack Incoming Webhook

Go to Slack API: Incoming Webhooks
Create a new webhook for your #support-escalations channel
Copy the webhook URL — you'll paste it into the Cowork task

Step 3: Create the Cowork task

Open Claude Desktop → Cowork tab → + New task.

Title: Flag repeat customer contacts and alert Slack

Description:

You are a customer support analyst. Your job is to identify customers who have
contacted support 3 or more times in the past 7 days, tag their latest ticket
in Gorgias, and post an alert to Slack.
 
Gorgias credentials:
- Domain: your-store
- Auth: HTTP Basic (username: you@company.com, password: YOUR_API_KEY)
 
Slack webhook: YOUR_SLACK_WEBHOOK_URL
 
Step 1 — Fetch recent tickets:
GET https://your-store.gorgias.com/api/tickets?limit=100&order_by=created_datetime:desc
 
Step 2 — Filter to tickets created in the last 7 days:
Compare each ticket's created_datetime to today minus 7 days. Discard older tickets.
 
Step 3 — Group remaining tickets by customer ID:
For each customer, collect their ticket count, email, name, and list of ticket
subjects with IDs.
 
Step 4 — Identify repeat contacts:
A repeat contact is any customer with 3 or more tickets in the 7-day window.
 
Step 5 — For each repeat contact, tag their most recent ticket:
PUT https://your-store.gorgias.com/api/tickets/{latest_ticket_id}
Body: {"tags": [{"name": "repeat-contact"}]}
 
Step 6 — Post a Slack alert for each repeat contact:
POST to the Slack webhook URL with this format:
{
  "text": ":rotating_light: *Repeat Contact Flagged*\n\n*[Customer Name]* ([email]) has opened *[count] tickets* in the last 7 days.\n\nRecent tickets:\n- #[id]: [subject] ([status])\n- #[id]: [subject] ([status])\n...\n\n<https://your-store.gorgias.com/app/ticket/[latest_id]|View in Gorgias>"
}
 
Step 7 — Print a summary:
List each flagged customer, their ticket count, and the ticket subjects.
If no repeat contacts found, print "No repeat contacts found — all clear."
 
Important:
- Only tag the MOST RECENT ticket per customer, not all of them
- Skip customers tagged as "high-volume-expected" (agencies, resellers)
- Count only customer-initiated tickets, not auto-replies or system messages

Customize the threshold in plain English

To change the threshold, just edit the task description. Replace "3 or more times in the past 7 days" with whatever makes sense for your business — Claude reads the description as natural language, so no code changes are needed.

Step 4: Set the schedule

Frequency: Every 2–4 hours during business hours

Repeat contact detection doesn't need to be real-time. A check every few hours catches patterns early enough for your team to act before the customer's frustration compounds.

Claude Desktop must be running

Cowork tasks only execute while Claude Desktop is open and your machine is awake. If you need overnight or weekend coverage, use the Agent Skill approach with a cron job instead.

Step 5: Run manually and review

Click Run now and watch the output. Claude will:

Fetch recent tickets from Gorgias
Group them by customer
Identify repeat contacts
Tag their latest ticket
Post Slack alerts
Print a summary

A typical run summary looks like:

Fetched 92 tickets from the last 7 days.
Grouped into 67 unique customers.
 
Found 3 repeat contacts:
 
Sarah M. (sarah@example.com) — 4 tickets in 7 days
  - #14501: Can't log in to my account (open)
  - #14487: Still can't access order history (open)
  - #14463: Password reset not working (closed)
  - #14451: Account access issue (closed)
  → Tagged #14501 as "repeat-contact" ✓
  → Slack alert sent ✓
 
James K. (james@example.com) — 3 tickets in 7 days
  - #14498: Wrong item received again (open)
  - #14472: Missing item in my order (closed)
  - #14460: Shipping damage (closed)
  → Tagged #14498 as "repeat-contact" ✓
  → Slack alert sent ✓
 
Maria L. (maria@example.com) — 3 tickets in 7 days
  - #14495: Coupon code not working (open)
  - #14478: Promo price not applied (closed)
  - #14466: Checkout error with discount (closed)
  → Tagged #14495 as "repeat-contact" ✓
  → Slack alert sent ✓
 
Done. 3 customers flagged, 3 Slack alerts sent.

Step 6: Brief your support team

Let your agents know what the repeat-contact tag means and how to use it:

Create a Gorgias View filtered to the repeat-contact tag — this becomes your escalation queue
When an agent picks up a flagged ticket, they should read all recent tickets from that customer before replying
The goal is root cause resolution, not just answering the latest question — look for the underlying pattern
If the pattern points to a product or process issue, escalate to the relevant team with a summary

When to use this approach

You want repeat contact detection set up in minutes with no code
Your ticket volume is moderate (under 200 tickets/day)
Checks every few hours are fast enough for your workflow
Your team is already using Claude Desktop

When to switch to another approach

You need real-time detection the moment a ticket is created → use n8n with a webhook
You need overnight or weekend coverage without a laptop running → use the Agent Skill with cron
You want to combine repeat contact detection with other automated actions (auto-assign, priority boost) → use n8n for the orchestration flexibility