CRM
Audit low-confidence Intercom topic labels into an Airtable review queue
Re-classifies stale Intercom conversations with OpenAI, and when confidence is low or topics conflict.
How it runs
The automated pipeline, trigger to output.
- TriggerSchedule starts the topic audit
- ActionPull stale-topic conversations from IntercomIntercom
- ActionRe-classify topic and confidence with OpenAIOpenAI
- LogicBranch on confidence and topic conflict
- OutputWrite confirmed and needs-review rows to AirtableAirtable
What it does
Automated topic labeling is wrong often enough that blindly writing to your CRM creates new bad data. This workflow re-classifies older Intercom conversations, and any case where OpenAI is uncertain or returns conflicting topics is parked in an Airtable review queue with the transcript, suggested topic, and confidence. A reviewer confirms or corrects it; only confirmed labels proceed.
When to use it
Use it when data quality matters more than speed, such as before a major segmentation rebuild, or when an earlier automated backfill produced labels you no longer trust.
How it works
A schedule kicks off the audit. The workflow pulls conversations whose topic label is missing or older than a threshold from Intercom. OpenAI returns a topic and confidence. A logic branch sends high-confidence labels straight to the confirmed table, while low-confidence or conflicting ones go to a `needs_review` Airtable table. The output step writes both buckets to Airtable so a human can clear the queue.
Set it up
What you configure once, before turning it on.
- 1Connect IntercomConversations, contacts, articles.
- 2Connect OpenAIModels, embeddings, files.
- 3Connect AirtableBases, tables, views, automations.
- 4Set each agent's modelWe leave models unset so you pick the tier — fast + cheap, or top-quality.
- 5Tune it to your dataEdit the prompts, filters, and field mappings so it matches how your team works.
- 6Test, then turn it onRun once against a sample, confirm the output, then enable the trigger.
More CRM workflows
Backfill Intercom conversation topics onto stale HubSpot contacts
Finds HubSpot contacts with no recent Intercom topic, pulls their conversation history from Intercom, classifies the dominant topic with OpenAI.
Daily orphaned-opportunity sweep with manager Slack digest
Each morning this scans Salesforce for open opportunities owned by inactive users and posts a per-manager digest in Slack listing the orphaned deals that need a new owner.
AI-Personalized Dormant Account Win-Back
On a dormancy anniversary, an agent researches the account's history and recent public signals, drafts a genuinely personalized win-back email.
Dormant Account Anniversary Re-engagement Email
Each day, finds Attio accounts whose last activity was exactly N months ago and sends each owner's contact a personalized re-engagement email from Gmail.
Dormant Account Check-In Call Scheduler
On the anniversary of an account's last touch, books a tentative check-in call on the owner's Google Calendar and emails the contact a scheduling invite.
Weekly Dormant-Anniversary Cohort Digest
Every week, groups dormant Attio accounts by which last-touch anniversary they hit that week, builds a prioritized digest.
Run it inside a business
This workflow drops into a full company template. Import the org, and this is one of the playbooks its agents run.
Run this workflow in your colony.
14-day trial. No DevOps. No Sales call. Provisioned in under a minute.