← All cases

AI-Powered Sales

AI Prospecting: The Agent Writes, the Human Sends

A system that builds the list, researches each business from public sources, writes a one-page brief with evidence, and prepares the message cadence. The first message never sounds like a bot — because a bot in the first message kills the conversation.

5
messages per prospect, with a 24h/+2d/+4d cadence
1 page
brief with a source for every claim
0
automatic sends — the finger on the button is human
Client
In-house operation (method distilled from 4 sources, in use since 08/2026)
Period
August/2026 — in operation, results not yet measured
Stack
Claude Code (skill) · Public sources (Maps, website, Instagram) · Cloudflare Pages · WhatsApp / email

Why this design

Anyone selling an “agent that books meetings on its own” is selling inbound. In outbound, automation that writes the cold first message becomes noise — the recipient doesn’t know you. So the system does everything except hit send.

What it delivers every morning

For each business on the list: a file with what was found (and where), a one-page brief on its customer service, and five ready-to-send messages — the first as a curious customer, the second asking for a yes-or-no on 15 minutes, three follow-ups with a new observation. A send checklist with times.

Rules

An unverified fact is --. Only public company data. It never sends, never posts.

Honest status

In operation since August 2026. No meeting numbers to show yet — when there are, they’ll go here. y’ll go here.