Methodology

How this works

A positioning audit that thinks like a PMM, not a chatbot. Built on a proven framework with an opinionated diagnosis layer on top.

01

Why positioning matters

Most B2B SaaS companies have a product that works. Their problem is not the product. Their problem is that the homepage does not explain why anyone should care. The headline hedges. The subhead lists features. The visitor leaves.

Positioning is the context you set around your product so the right buyer immediately understands why it matters to them. Get it wrong and your marketing, sales, and product messaging all drift in different directions.

02

Why April Dunford's framework

April Dunford's framework from Obviously Awesome is the most widely adopted positioning model in B2B SaaS. It breaks positioning into five components: competitive alternatives, unique attributes, value, best-fit customers, and market category.

This tool uses these five components as the structural backbone of every audit. They force specificity. You cannot hedge when you have to name the competitive alternative. You cannot hand-wave when you have to connect an attribute to a value the customer cares about.

The framework is the lens. But filling in five boxes is not an audit. That is where the diagnosis comes in.

03

The diagnosis layer

The part that makes this more than a framework exercise. The diagnosis reads the homepage copy as a PMM would and calls out three things:

  • Hedging language.Words and phrases that sound good but say nothing. "Modern," "streamline," "best-in-class." If you could swap in any competitor's name and the sentence still works, it is hedging.
  • Contradictions. When the homepage promises simplicity but the product requires setup. When the hero targets developers but the pricing targets enterprise. Internal contradictions that confuse the buyer.
  • What is missing. The proof point that is not there. The competitor comparison that is avoided. The performance claim that would be convincing but is never made.

The diagnosis ends with the single highest-leverage change. Not a list of ten suggestions. One thing. The one change that would do the most to sharpen the positioning.

04

Why not just ask ChatGPT

You could paste a homepage into any LLM and ask for positioning feedback. You will get a polite, balanced, hedge-everything response that reads like a consulting deck. It will tell you your positioning is "strong but could be clearer." It will not commit to what is actually wrong.

This tool is different because the prompt engineering is opinionated. The system prompt bans filler language, forces specific recommendations, and requires a concrete rewrite, not a description of what a better one might look like. The output reads like a PMM talking to a founder, not a chatbot summarizing a webpage.

05

How BYOK works

BYOK stands for "bring your own key." You provide your Anthropic API key to run an audit. The key is held in your browser, sent over HTTPS for a single server-side API call, and immediately discarded. It is never stored, logged, or transmitted anywhere else.

This means you pay Anthropic directly for the API usage (roughly $0.05 to $0.10 per audit) and this tool never handles billing, subscriptions, or your credentials. The showcase audits on the homepage are pre-generated and cost nothing to view.

06

Who built this

I'm Teddy Castro, an AI-native full-stack product marketing manager. I build tools that show how AI changes PMM work, not by replacing the thinking but by making the analysis faster and more rigorous.

This tool is built with Next.js, TypeScript, and Claude. The source is open on GitHub.

See it in action

Browse the showcase audits or run your own.