I've been using tl;dv for about six months to record and transcribe all our sales demos, and the library is getting... massive. We're easily past a thousand calls now. The search is great, but sometimes you need to find *a type* of call, not just one with a specific keyword.
My current system is breaking down. I started with simple folders by quarter (`Q1-2024`, `Q2-2024`), but that's just a chronological dump. I then tried tags like `#competitor-A-mention` or `#pricing-question`, but it's become a tag explosion. I'm not consistent, and my team uses different tag names for the same thing (#objection vs #price-objection).
I'm curious how others are structuring this. For a sales team, what's the scalable way?
* **Folders by ICP, deal stage, or feature?** (e.g., `Enterprise/Discovery`, `SMB/Product-Demo`)
* **A rigid tag taxonomy?** If so, how do you enforce it? Do you use a naming convention like `competitor:competitor-A`, `topic:pricing`?
* **Combination approach?** Folders for broad segments, tags for granular details.
Also, has anyone used the API or webhooks to auto-organize recordings? I'm thinking of a Zapier/Make workflow that parses the transcript or meeting title and applies tags/folders automatically. The API docs show you can add topics (which I think are tags) programmatically.
```json
// Example from their API docs for adding topics
{
"recordingId": "rec_123",
"topics": ["competitor-A", "feature-x-demo"]
}
```
This could be a game-changer for consistency. Imagine tagging calls based on transcript analysis for "integration" or "security question" mentions.
What's working (or not working) for you? Any pro tips for managing a huge call library?
Webhooks or bust.