Prompt of the day

Segment hypotheses from the data you already have

Everyone wants personas; you have a CRM export and campaign history. Build evidence-based segment hypotheses before anyone invents imaginary people.

You are an audience strategist who builds segments from live data, not stock personas. I will describe the fields I actually have and the behavioral differences I have observed. Produce:

A) A SEGMENT HYPOTHESIS TABLE: segment name, defining criteria using only fields I listed, the evidence for it in my data versus what is still assumption (two separate columns), and estimated size — writing "size not provided — pull the count" where I gave none.
B) For the top THREE segments: the message angle each one's evidence supports, and the channel where my data says they are actually reachable.
C) A VALIDATION PLAN: the cheapest test per segment hypothesis, existing-data reanalysis before any live spend, with what a confirming and a killing result each look like.

My data: [DESCRIBE: CRM/ESP fields available, engagement and purchase patterns observed, acquisition sources, anything known about high-value customers]

Rules: Do not invent, estimate, or fabricate any statistic, benchmark, or performance figure — if a number is not in the material I give you, write "not provided" and flag it. Mark every claim I should verify in my analytics or source systems before it is published or presented. Never include customer personally identifiable information or client-confidential terms.

Why it works — Persona decks invent people; segment hypotheses built from fields you can actually target ship campaigns. The evidence-vs-assumption split is the honesty device — it shows exactly how much of the team's 'audience understanding' is still fiction before budget follows it.

Rehearse a hard conversation with AI first

The raise, the pushback, the tough client call — rehearsing once against a realistic counterpart beats winging it live.

Say: 'Play [my manager / this client]. I'll practice asking for [X]. Push back realistically, then tell me how I did and what to say instead.'

1. Set the scene and who AI is playing.
2. Run the conversation for real.
3. Ask for feedback + better lines.