Outbound AI Lead Qualification Agent: the thinking.
An outbound voice agent that calls a real-estate lead, introduces the service, qualifies them on budget, timeline and property preferences, and writes every answer to a sheet.
The starting point.
What made this worth building.
Real-estate leads go cold fast, and the first call is always the same handful of qualifying questions. Doing them by hand means slow follow-up, and answers that live in someone’s notes instead of a shared record.
The direction I chose.
The approach taken.
A Retell voice agent driven by n8n. It places the call, introduces the service, asks the qualifying questions — budget, timeline, property preferences — and writes the answers into Google Sheets. Conditional follow-up and callback flags are wired end to end, so a lead who asks to be called back is marked for it.
How the pieces connect.
Each stage exists because the one before it cannot be trusted on its own.
- 01Lead list
- 02Outbound call
- 03Introduce the service
- 04Qualify: budget · timeline · preferences
- 05Answers to Google Sheets
- 06Follow-up / callback flag
The decisions behind it.
The constraints I set deliberately, and would defend.
- Ask one qualifying question at a time, the way a human ISA would, rather than reading a script.
- Write structured answers to a sheet as the call happens, so qualification is a record and not a recollection.
- Flag callbacks and follow-ups explicitly instead of leaving them to whoever reads the notes.
What has to hold true.
How I would know it works — and how I would know it does not.
- Validated with live test calls end to end: call placed, questions asked, answers written, callback flags set.
- Never run against real leads.
Where this stands today.
Stated plainly, with nothing implied.
This project is complete.
Designed and built the agent and its n8n workflows. Personal prototype, validated with live test calls only — not a client or production deployment.
Keep going