Thousands of people walk through your page before your customers do.
Pick a crowd: skeptical founders, busy buyers, first-time shoppers on a phone. Your own coding agent walks your page as each of them and comes back with what they would say: would they sign up, what confused them, what they did not trust.
Illustration: a queue of little people walks into a big browser window labelled "your page" and comes out the other side with speech bubbles.
Step one · one line
Add it to the tool you already use
dataready runs on your computer: a small MCP server your coding agent starts by itself, inside your app's folder. Pick your tool, copy the line, sign in with your email once. Your app and everything we learn about it stay on your machine; we keep the personas and the know-how.
Packagenpx -y dataready-mcp
Claude Codein your app's folder
claude mcp add dataready -- npx -y dataready-mcp
Add -s project to write .mcp.json for your whole team. Then say map my app; the first time, it opens a sign-in page.
MCP Servers → Configure → paste, with your app's folder after --root.
On Windows, use "command": "cmd" with "args": ["/c", "npx", "-y", "dataready-mcp", …]. Needs Node.js 22.13 or newer.
Only want a quick check of a public page, without installing anything? Connect the quick-check server as a connector instead (Claude Desktop / claude.ai: Settings → Connectors → Add custom connector): https://mcp.dataready.si/mcp
How it works
Three steps. About as long as a coffee.
Install
Add one line to Claude Code, Cursor, Codex, Windsurf, Cline or Claude Desktop. Sign in with your email. Done.
Say who should look
Ask in plain words. We pick matching people from the crowd and hand your coding agent a short brief for each one.
test my page with 20 EU founders
Read the report
Your agent walks the page as each person and sends back what they saw. You get one page: who would sign up, why the rest would not, and where it broke.
dataready runs on your computer and your page is opened by your coding agent, with its own browser. Your code and your results never leave your machine. No browser in your tool? We can read the public page for it.
What you learn
The four questions every launch has
?!
Would they buy?
Each person scores how likely they are to sign up today, and says yes, maybe or no to your main button, with the reason in their own words.
Aa
Is the copy clear?
After five seconds, what do they think it is? Which words lost them? Who did your headline speak to, and who did it leave out?
!
Where does it break?
A friction map of the steps that stopped people, from a broken button to a form that asks too much too soon.
ok
What do they distrust?
The proof they looked for and did not find: reviews, a real company behind it, privacy, a clear way to leave.
Sample report
One page you can act on before lunch
Twenty skeptical EU founders on a fictional project-management landing page. Every report is private, behind an access code, and gone after seven days.
example-app.com · 20 personas · EU SaaS founders, skepticalSample
C
64 / 100 overall
confidence: high · 20 walks
52would sign up
71copy
58trust
90it works
Would take the main action now
yes 25%maybe 40%no 35%
Friction map
blockerSign-up form asks for a company size and phone number before showing anything. 6 of 20 stopped here.
majorPlans are only visible after creating an account.
minorHeadline "the operating system for work" did not say what it does.
In their words
"Looks polished, but I could not find who runs it. No address, no team. I'd leave here."Skeptic · founder · DACH · desktop
"The demo video sold me in thirty seconds. I'd start the free plan today."Early adopter · founder · Nordics · mobile
The people are simulated. Treat their answers as the fastest first opinion you can get, not as a substitute for real customers.
The crowd
Over 2,400 people, each with their own story
A procurement head who signs off alone. A student on a cracked phone. A CFO who needs proof before believing a claim. Each one has a role, a market, a budget, habits and pet peeves. Browse them, filter them, or add your own.
prodready makes what you built with Lovable, Replit, Cursor or Claude Code safe to put in front of real users. dataready tells you whether those users will want it. Same install, same sign-in style, made by the same team.
Yes. There is no paid plan and nothing to upgrade to. We built it because our own customers kept asking "would anyone actually buy this?".
Are these real people?
No. They are simulated people with detailed, consistent profiles, played by your own coding agent. They give you a fast, honest first opinion. They do not replace talking to real customers, and the report says so.
Do you see my code or my computer?
No. dataready runs on your computer, reads only your app's folder, and keeps what it finds in .dataready/ inside your project. Nothing of it is sent to us. From us it downloads the personas and the know-how: how to read an app, how people behave, how to score them.
My tool has no browser. Does it still work?
Yes. Ask for a text snapshot and we read the public page once and put its text into the brief. Pages behind a login cannot be read this way.
Will it fill in my forms or pay for things?
Never with real data. The brief tells your agent to use a test address, to stop at any payment form and never to log in with real credentials.
What do you keep?
Your sign-in email. Your app's profile and its simulated community stay on your computer. For a quick check of a public page through the quick-check server, we keep the results for seven days so the report page works. The tested address is kept without its query string. Details are on the privacy page.
Can I add my own persona?
Yes, anyone can. Propose one on the site or ask your coding agent to "propose an agent". We review every proposal before it joins the crowd.
dataready
