Paul Nugent — Product Marketing + AI

PMMstack

Four skills for Claude — from URL to positioning

Most AI tools in product marketing are one-shot: paste your idea, get a report. The AI does the thinking for you. But where's the fun in that? These are different. One is a fast research tool, three run as structured conversations that force you to think first, then use AI to validate, challenge, and extend your thinking. Together they take you from a product URL to a complete positioning document. (Yes, this approach is inspired/stolen/nicked outright from the gstack /office-hours skill which I've found very helpful.)

JTBD Miner

Generate ideas for jobs to be done in your ecosystem

Give it any product or company URL. It fetches the page and analyses it through six lenses — feature-to-outcome translation, customer voice, case study headlines, pricing structure, audience signals, and conspicuous absence. Produces 3-5 candidate job statements with signal strength ratings. Use your own URL or a competitor. The fast entry point to the chain.

Demand-side analysis JTBD extraction

feeds into

JTBD Validator

Is this job real, underserved, and worth building against?

Five-phase session that sharpens your job statement using Christensen's definition tests, walks you through Moesta's Four Forces, then searches for external evidence to validate or challenge your assumptions. Produces an opportunity score grounded in real signals, not vibes. Phases 1-2 are blocking — you articulate what you know before the AI researches.

Christensen Ulwick Moesta Klement

feeds into

ICP Definer

Who has this job most acutely?

Takes your validated job and helps you generate 2-3 candidate ICPs defined by circumstance, not firmographics. Scores each on nine dimensions — job intensity, switching readiness, willingness to pay, success potential — with bottom-up market sizing. Produces ICP Cards you can take into strategy sessions or positioning work.

Murphy Vohra / Ellis Dunford Moesta

feeds into

Positioning Workshop

How should you position for this customer?

A guided workshop based on April Dunford's Obviously Awesome methodology. Walks through competitive alternatives, unique attributes, value themes, target customer, and market category — with firm pushback on weak answers drawn from Dunford's actual correction patterns (I used podcast transcripts, blog posts, etc of Dunford's voice). Produces a structured positioning document with the full reasoning chain.

Dunford Obviously Awesome Sales Pitch

How to use these

PMMstack skills run inside Claude (Anthropic's AI). They work in Claude Code or Cowork mode.

1 Clone the repo: git clone github.com/paulnugentuk/pmmstack
2 Copy the skill folders into ~/.claude/skills/
3 Start a Claude session and the skills will trigger when you mention JTBD mining, validation, ICP definition, or positioning work

# Quick install
git clone https://github.com/paulnugentuk/pmmstack.git
cp -r pmmstack/jtbd-miner ~/.claude/skills/
cp -r pmmstack/jtbd-validator ~/.claude/skills/
cp -r pmmstack/icp-definer ~/.claude/skills/
cp -r pmmstack/positioning-workshop ~/.claude/skills/

About

The short version

I'm a product and growth marketer by training — I've spent my career thinking about positioning, buyer psychology, and how insights actually translate to impact. Now I'm exploring what happens when you give AI the same frameworks that good PMMs use and structure the interaction so the human still does the thinking work.

PMMstack is the first public output of that exploration. These aren't chatbots that generate positioning for you — they're structured facilitation tools that push back on weak thinking, force specificity, and produce outputs grounded in real evidence. The same approach a good consultant would take, encoded as repeatable AI skills.

I'm available for Product Marketing consulting work. Hit me up on Linkedin.