CRM

Escalate Churn-Risk Intercom Chats to a Slack Save Channel

Detects cancellation and frustration language in live Intercom conversations and posts a structured escalation to a Slack retention channel with the customer, the quote…

CategoryCRM
Enginesim
Difficultybeginner
Triggerevent
Steps5
Setup~5 min

How it runs

The automated pipeline, trigger to output.

  • TriggerNew inbound Intercom messageIntercomIntercom
  • ActionFetch recent thread contextIntercomIntercom
  • ActionScore churn risk + extract key quoteOpenAI
  • LogicContinue only on high risk
  • OutputPost escalation to Slack save channelSlack

What it does

This workflow watches incoming Intercom messages for churn-risk language — cancellation requests, pricing complaints, repeated bug reports, or threats to leave. When it finds one, it posts an alert to a dedicated Slack save channel containing the customer name, plan, the exact at-risk quote, the classifier's reasoning, and a one-click link back to the Intercom conversation so a CS lead can intervene before the chat goes cold.

When to use it

Use it when minutes matter for retention and you want a human on the highest-risk conversations while the customer is still typing. Best for teams that triage saves in Slack rather than a queue.

How it works

  1. 1A new inbound Intercom message triggers the workflow.
  2. 2The recent thread context is fetched so the classifier sees the whole exchange, not one line.
  3. 3An LLM scores churn risk and extracts the single most concerning quote.
  4. 4A branch only continues when risk is high, so low-stakes chats never page anyone.
  5. 5A formatted alert with customer, plan, quote, and a deep link posts to the Slack save channel as the final output.

Set it up

What you configure once, before turning it on.

  1. 1
    Connect IntercomConversations, contacts, articles.
  2. 2
    Connect OpenAIModels, embeddings, files.
  3. 3
    Connect SlackChannels, DMs, threads, mentions.
  4. 4
    Set each agent's modelWe leave models unset so you pick the tier — fast + cheap, or top-quality.
  5. 5
    Tune it to your dataEdit the prompts, filters, and field mappings so it matches how your team works.
  6. 6
    Test, then turn it onRun once against a sample, confirm the output, then enable the trigger.

Run this workflow in your colony.

14-day trial. No DevOps. No Sales call. Provisioned in under a minute.