Complete AI Training
Sign inGet my AI kit

Your job's AI kit

Get your AI kit

Tell us who you are and what you do. We show you your kit right away and email you the link: skills, prompts, AI agents, MCP servers and courses for your job.

500+ jobs ready, and we make a kit for any other job. No payment needed to look.

Share

AI agent for product marketing managers

Persona Evidence Refresh Agent

Give each persona a current version in which every claim is backed by data.

CreatingMarketing
See how it worksGet access
Persona Evidence Refresh Agent: what goes in, what the agent does and what you get

What it does

A persona written three years ago describes 'Mid-market Mike' who no longer buys the product the way it did. The agent takes each persona claim, such as role, goals, pain points, buying triggers or channels, and tests it against current data from the CRM, support tickets and survey results. It marks each claim as supported, weak or contradicted, and shows the numbers and examples behind each. When the evidence for a claim is weak, it goes back to gather more, for instance more closed deals or more survey rows, before proposing a change. It then drafts revised persona text with the evidence next to each change, and highlights new traits that appear in the data but not in the persona. Edge case: a segment too small to support any conclusion. The manager approves persona changes.

How it works

Follow the arrows from top to bottom. The orange dashed arrow is the loop: when a check fails, the agent goes back and tries again.

Start and resultWhat it doesA check on its own workWaits for your OKGoes back and retries
Yes, continueYes, continueApprovedNoNo 1 STARTS WHEN Persona review starts 2 DOES Split each persona into separate claims 3 USES A TOOL Pull CRM, support and survey data for the persona'ssegment 4 DOES Test each claim against the data 5 CHECKS THE RESULT Does each claim have at least 30 relevant records? If not: widen the date range or segment and gather moredata. Back to step 3. 6 DOES Mark each claim supported, weak or contradicted 7 DOES Find traits in the data that the persona does notmention 8 DOES Draft the revised persona with evidence beside eachchange 9 CHECKS THE RESULT Does every changed line cite its evidence? If not: add the source and numbers or remove the change.Back to step 8. 10 YOU APPROVE Manager approves persona changes 11 RESULT Updated personas published
Read the steps as a list
  1. Persona review starts
  2. Split each persona into separate claims
  3. Pull CRM, support and survey data for the persona's segment
  4. Test each claim against the data
  5. Does each claim have at least 30 relevant records?If not: widen the date range or segment and gather more data. Back to step 3.
  6. Mark each claim supported, weak or contradicted
  7. Find traits in the data that the persona does not mention
  8. Draft the revised persona with evidence beside each change
  9. Does every changed line cite its evidence?If not: add the source and numbers or remove the change. Back to step 8.
  10. Manager approves persona changesThe agent waits here for your OK.
  11. Updated personas published

How it decides

A claim is supported with at least 30 relevant records and a clear pattern. Claims with fewer records are weak, and more data is gathered before change.

  • Require 30 records before judging a claim
  • Mark a claim contradicted when data goes against it in 60 percent of records
  • Propose a new trait when it appears in 25 percent of records
  • Keep the persona's original wording where the data supports it

Make it yours

Every agent is a starting point. You choose these settings for your own situation.

  • Personas included
  • Minimum records (default 30)
  • Data sources
  • Review schedule
  • Evidence format

What keeps you in control

It always asks you first

  • Manager approves persona changes
  • Manager approves sharing personas externally

Hard limits

  • Never use customer names in public persona text
  • Show evidence for each claim

It stops when

  • Done: personas updated with evidence
  • Stop: CRM data is missing key fields, list what is needed

Set it up

We guide you through the set-up, step by step

Members get the full set-up guide for this agent. No technical skills needed: you copy, paste and upload.

10 minto set it up in your AI
5 AIsChatGPT, Claude, Copilot, Gemini, Grok
  • One set of instructions to paste into your AI, with the clicks for ChatGPT, Claude, Microsoft 365 Copilot, Gemini and Grok
  • The agent then walks you through connecting your own data, one source at a time
  • A downloadable copy with the flow chart, the rules and the full guide
Get access to this agent

An example run

What happensFor 'Ops Olivia', the claim 'buys through a security review' is tested on 52 closed deals and holds in 41, 79 percent. The claim 'prefers phone demos' holds in 11 of 40, so it is contradicted. Survey data shows self-serve trials in 38 percent of deals, a new trait. A first draft cites only 22 records for one claim, so the agent widens the range to 46. The manager approves.

More agents for product marketing managers

Research · Quality control

Competitive Claim Verification Agent

A competitive brief where every claim is dated, sourced and fairly compared.

7 steps1 check14 jobs
Quality control · Product development

Product Pricing Page Consistency Agent

Pricing pages that match approved packaging.

8 steps2 checks1 job
Quality control · Sales

Sales Competitive Response Review Agent

Fair, sourced competitive responses.

8 steps2 checks8 jobs
Communication · Automation

Campaign Asset Approval Chase Agent

Every campaign asset is approved before launch, or a fallback is ready.

9 steps1 check4 jobs