Outbound system: cold outreach run like operations
A cold outreach machine that ran 65 days straight: 1,400+ individually personalized DMs, a verified ~5% reply rate, every send logged, and no machine ever replying to a human on its own.
outreach · automation · human-in-the-loop · evidence-logging
A cold outreach machine that ran every single day for 65 days: 1,400+ individually personalized DMs, a verified ~5% reply rate, a permanent log of every message sent, and not a single message ever sent to a human by a machine acting alone.
The pipeline: a 3,000-row scored prospect list built from public profiles, messages drafted one person at a time against each prospect’s public profile, daily volume that paces itself from its own track record, and a human-operated reply desk behind three content gates.
The three disciplines that made it work:
- It runs every day, and I know it ran. Volume grows on clean streaks and adjusts down the moment the data says to. The system reports in daily with a 30-second summary, and the absence of that summary is itself the loudest alarm: a dead system cannot announce its own death.
- Measure what actually happened, not what an API claims. The platform’s official API reported 6 replies where 41 actually existed, a 7x undercount. Every reply is verified against what a human eye would actually see on the page, automatically, five minutes a day.
- Machines draft, humans converse. Message variants ran as controlled experiments (no verdict before 150 sends, anything under 2% killed), and no machine ever replies to a human on its own: a human approves every single send.
The full writeup, with the numbers and the standards, is the flagship build in 15 Days with Fable.
If you have a product people already love and you want outbound run against it with this level of rigor: message me at @phuakuanyu. I will walk you through live logs and dashboards from a real operation before you commit to anything.