// Case study
AI analysis of sales calls
Call recording → transcript → AI summary, sentiment and next steps in a sheet. The manager gets recommendations in Telegram, the head gets an email about negative calls.
// Task
The head of sales can't listen to every call
Managers make dozens of calls, and the head finds out about a problem client when it is already too late. There is no time to re-listen to recordings by hand.
// Solution
Every call analysed a minute after upload
A new recording in a Google Drive folder triggers the scenario: AssemblyAI transcribes it, and AI reviews it like a sales analyst — who is speaking, what was agreed, the sentiment and what to do next. Everything goes into a calls sheet.
The manager gets a summary with concrete recommendations in Telegram. If the call went badly, the head gets an email right away.
- Audio transcription with speaker detection
- Summary, sentiment and next steps for the CRM
- Recommendations to the manager in Telegram
- Email to the head only about negative calls
// Architecture
How it works
// Result
The outcome
// Want the same?
Need call quality control?
Tell me where your recordings are stored and what matters to track, and I'll show you how to automate it.