
The product development system for teams and agents
UX Teardowns
Every capture is broken down into the patterns it uses and scored 1–5 against a published design rubric, with each score anchored to something actually on the page. Take the patterns into your next design review, or the scores into the conversation about why your own pricing tiers aren’t landing.
Linear
Figma
OpenAI
Netflix
Ramp
Cloudflare
Miro
PostHog
Slack
Dropbox
Coinbase
Airtable
Stripe
Vercel
Airbnb
Spotify
Cursor
Duolingo
Monday.com
Retool
Discord
Zapier
Robinhood
Atlassian
Notion
Anthropic
Shopify
Canva
GitHub
Intercom
Perplexity
Supabase
Reddit
HubSpot
Asana
UberHomepages, pricing pages, AI feature flows and search experiences. Every one broken into the patterns it uses and scored against the published rubric — searchable by pattern, so you can see everyone who has already tried the thing you are considering.
homepage and pricing experiences captured and scored
categories: homepage, pricing, AI features and search
per dimension, each score evidenced against the page
The platform
Structured examples of homepage, pricing, AI feature, and search experiences from leading SaaS and AI products — tagged by pattern and scored.
260+ teardowns loggedShowing 4 of 4 · select a capture to see how it scored
UX Scan
UX Scan is the library pointed at a page of your choosing. Paste a URL — your product, a competitor, something you shipped last week — or upload a screenshot. It captures the full page, works out whether it is a homepage, a pricing page or something else, and scores it on the rubric that matches. What comes back is a report: a score per dimension with the evidence behind it, and your position against every page already in the library.
Homepages and pricing pages are benchmarked against every tracked company. Anything else gets a general UX health check.
Capturing page…
Two ways in, and it works out the rest. A URL is captured for you; an image is scored as-is. Page type is detected automatically and you can override it, because the type decides which rubric runs — and a mis-detected page is the one thing that makes a scan read oddly.
Insights
Every page that gets scored rolls up into a view of the market: which homepage approach actually dominates, how far AI has moved up the page, what teams typically score on each dimension, and how pricing is structured underneath it.
Archetype assigned per capture with a confidence rating, from the same rubric every homepage is scored against.
Product-led wins by a distance — but it is not the default. Nearly half of them lead with the product itself, which is worth knowing before you commit to a hero built around a promise instead of a screenshot.
Ask the library a design question and get the range, not one example. Because every page is captured and scored the same way, a question about how something is designed comes back as the spread of approaches actually in use — including the pages that do not do it at all.
The rubric
Each kind of page is scored against its own dimensions — a homepage on four, a pricing page on three, anything else on a general set — and every score has a written definition behind it, so a 2 and a 5 mean the same thing on every page judged the same way. Pick a dimension and move the score to see what the rubric is actually looking at.
Four dimensions. Every tracked homepage is scored on these.
Whether the hero tells you what this is, who it is for, and why it is different.
A slogan and nothing else. No category, no audience, no claim.
Cannot tell what the company does after reading the hero
A growing library

The product development system for teams and agents

Transparent per-transaction pricing, no sales call

An agent that works on the canvas, not in a sidebar

Search by mood, act on the result immediately

Shift your product into self-driving mode

Hire, manage, pay, & equip anyone, anywhere.
See more of the platform