Day 8: Pick three real candidates

Three orgs from your own pipeline — one boring, one obvious, one wildcard.

Yesterday

Week 1 closed yesterday with the composite. Week 2 starts now — and the only way to lock in the framework is to run it.

Why this signal matters

Reading the framework is theory. The skill comes from running it. Picking three candidates with deliberately different profiles trains you to see what the composite catches that your gut misses, and where your gut still beats the score.

The 5-minute exercise

1Open your existing pipeline (CRM, notes app, Notion, whatever).
2Pick #1 — the boring one. A startup you'd assume scores 3/6, no obvious tells.
3Pick #2 — the obvious one. The startup you already feel good about.
4Pick #3 — the wildcard. A pre-revenue or stealth-ish org you can't quite read.
5Note today: GitHub URL for each, a one-line gut prediction (score 0-6), the reason.

What you’re filtering for

Three orgs that are public on GitHub. If a candidate has no public repos, pick again — the framework is GitHub-native and we want a clean comparison.

Edge case

If your pipeline is empty, pick three Series A/B announces from the last 30 days of TechCrunch. The framework works retroactively too — you'll see what the headline missed.

Bonus

Write down the gut prediction BEFORE you score. The most useful artifact of the week is the delta between your gut and the composite — both are data.

Tomorrow

Tomorrow: candidate #1, end-to-end, in 15 minutes. We'll do all six signals on the boring one first.

Curriculum: /challenge · Methodology: /methodology · Paper: ssrn.com/abstract=6606558