PAST Framework Quick Guide

PAST Framework Quick Guide

Most AI projects fail because of unclear strategy, not bad technology. Four questions — answered honestly — determine whether a project succeeds before you choose a tool.

PAST stands for Purpose, Audience, Scope, and Tone. Work through all four before you open a single product page or trial signup.

---

Purpose — Why are you using AI?

The wrong question: “How can we use AI?”

The right question: “What business outcome do I need?”

Starting with the technology and working backwards to value creates solutions looking for problems. Flip it. Start with the outcome you want, then find the AI that serves it.

The Metric Mandate: If you can’t put a number on your goal, you don’t have a goal — you have a wish. Before any AI project moves forward, complete this sentence:

“Success means [measurable outcome] by [timeline] as measured by [metric].”

If any part of that sentence is blank or vague, the project isn’t ready. Return to Purpose.

---

Audience — Who is this for?

AI implementations fail when they’re designed for the person buying the tool rather than the person using it. Two questions to answer before you go further:

Primary users: Who interacts with AI daily? What does their current workflow look like, and where does it frustrate them?

Output consumers: Who sees the results? A technical team reading code documentation needs different output than an executive reading a summary — and both need different output than a client reading a proposal.

The test: does this make their job easier or harder? If you can’t answer that with confidence, you don’t know your audience well enough yet.

---

Scope — What’s in and out?

Scope creep kills more AI projects than technical failures. Start narrow, prove value, then expand.

Define what AI handles. Which tasks, which data sources, which decisions can AI make versus which require human approval?

Define what stays human. Writing this down is as important as the first list. Without explicit boundaries, scope drifts — and drifting scope means diffuse results.

Start with Phase 1: One use case, limited users, specific data. Expand after you’ve demonstrated the model works.

The research backs this up: AI leaders pursue half as many opportunities as their peers but scale twice as many successfully. Tight scope isn’t a limitation — it’s how you get things across the finish line.

---

Tone — How should AI sound?

AI implementations succeed when the output sounds like something you’d actually say. They fail when every email, proposal, and post reads like it came from the same generic robot.

Match your actual communication style, not a generic “professional” default. If your client emails are direct and casual, train for that. If your proposals are structured and formal, build for that.

The AI slop test: Read your AI-generated output and ask: “Could this exact text have been written by anyone in my industry?” If the answer is yes, rewrite it with specifics. Your industry knowledge, your client context, your particular way of framing things — that’s your competitive advantage. AI should preserve it, not flatten it.

---

Quick Reference

Element	Question	Red flag
Purpose	What measurable outcome?	”Use AI more” (no number)
Audience	Who uses it and who sees the output?	”Everyone” (too broad)
Scope	What’s in and what’s out?	”Everything” (no boundaries)
Tone	How should it sound?	”Professional” (too vague)

---

Apply PAST before choosing any tool. If any element is blank, the project isn’t ready.

Ready to work through it? The PAST Wizard walks you through Purpose, Audience, Scope, and Tone for your specific situation.