Hey everyone, I've been refining my meeting analytics workflow for the past quarter and wanted to share my current setup. It's been a game-changer for getting a data-driven pulse on our marketing syncs and customer calls.
My core stack is **Fireflies.ai -> Airtable -> Looker Studio** (formerly Data Studio). Here’s the basic flow:
1. **Fireflies.ai** handles the capture and initial transcription. I have it integrated with our team's Google Calendar, so it auto-joins and records.
2. The magic is in the **Fireflies -> Airtable integration**. After each meeting, Fireflies sends the transcript, summary, action items, and even custom metrics I track (like "mentions of competitor X") into a dedicated Airtable base.
3. In **Airtable**, I do some light data massaging. I add fields for:
* Meeting Type (e.g., "Marketing Standup," "Client Onboarding")
* Department Owner
* Key Decision Tags
* A manual "Sentiment" score (1-5) based on my read of the transcript.
4. **Looker Studio** then pulls directly from Airtable. I've built a weekly dashboard that shows:
* Total meeting hours analyzed
* Action items by department & completion status (from Airtable)
* Trends in discussed topics (using word frequency from transcripts)
* Average sentiment by meeting type over time.
The biggest benefit has been spotting recurring topics in customer calls that we weren't addressing in our nurture streams. It also keeps us honest on action items!
**Pitfalls to note:**
* The Airtable integration can sometimes duplicate entries if a meeting link is shared twice. I have a simple automation to flag potential duplicates.
* You need to be disciplined about tagging meeting types in Fireflies or Airtable for the segmentation to work.
* The sentiment scoring is manual on my part—I haven't found an automated way I trust yet.
Has anyone else built something similar? I'm curious about alternative ways to handle the sentiment analysis piece, or if you're using different connectors.
Cheers, Henry