00TriggerBuild 22 of 30
Voice agents · Sales team
AI Sales Call Summary and CRM Logging System
Stack
- Vapi
- HubSpot
- OpenAI
- n8n
- 01 · VapiCall ends
- 02 · n8nTranscript
- 03 · OpenAIStructured summary
- 04 · HubSpotUpdate deal and task
01InputThe problem
The problem
Call notes were inconsistent or missing, so managers could not coach and deals slipped because next steps were forgotten.
02ReadDiscovery
Discovery
Before building, I mapped:
- Q1What a good call summary contains for this team
- Q2Which HubSpot fields should update after a call
- Q3How follow-ups are assigned
03RouterThe idea
The idea
Record and transcribe the call, let AI write a summary in the team’s format, and push it to HubSpot with the next step already scheduled.
04IteratorHow it runs
How it runs
- Step 1
When a call ends, the recording and transcript are collected.
- Step 2
OpenAI writes the summary: needs, objections, budget, next step.
- Step 3
n8n updates the HubSpot deal and contact fields.
- Step 4
A follow-up task is created for the rep with the agreed date.
05OutputHandover
What the client got
- n8n workflows
- Summary template and prompt
- HubSpot field mapping
- Manager view
06RouteNext build
Next build
Next build · Workflow automationAI Social Media Content Generator for Brands07ActionBook a call
Hand me the busywork.
An automation, a GoHighLevel setup, an AI agent, a connected system or a batch of AI ads. Tell me what slows you down and I will tell you how I would fix it.
