mocktomer

FAQ

The things you'd ask before pasting your URL

Straight answers on safety, money, and whether the feedback is real. Anything missing? Email us — a human replies.

Will agents place real orders or payments on my live site?
Never. It's a hard guardrail, not a setting: agents test your checkout and forms up to the point of payment, then stop and report how it felt. They never submit real payment data or complete a real order. Full end-to-end checkouts happen only on your staging environment, with test cards you provide.
Is this QA or bug testing?
No — your test suite checks whether the site works; mocktomer checks how it feels to use. The findings are experiences ("shipping felt like a trap", "I couldn't find the free plan"), not stack traces. It complements QA; it doesn't replace it.
Do shoppers check accessibility?
Both, now. Shoppers feel it — unlabeled buttons, keyboard traps, and unreadable contrast land in your report in their own words — and every page they visit also gets a formal WCAG A/AA scan, so the report lists the exact rule violations, worst first, with links on how to fix each one. There's also Ellis, a dedicated access-first shopper who navigates keyboard-first; add them to any run from your persona list.
Is AI feedback actually legit?
Judge it yourself — that's the point of the free tier. Every finding ships with the full journey transcript: screenshots plus the customer's step-by-step reasoning. Nothing is a black box, and if a claim doesn't hold up you mark it Ignore and it never nags you again.
How is this different from Hotjar or analytics?
Analytics and session replays show you where real visitors dropped off — after it already cost you. mocktomer's customers narrate why a moment felt wrong, in words, and they'll do it before real visitors ever hit the problem. Use both: they answer different questions.
Do I need to install anything?
No scripts, no snippet, no code on your site. Paste your URL, confirm it's yours, and we start crawling in about two minutes — nothing to install, nothing to verify.
Can I share results with my team or a client?
Yes — every completed run can mint a public replay link: anyone with the link watches the shoppers' journeys (screenshots plus their reasoning) without an account, and you can revoke it in one click. Reports also export as branded PDFs.
Can I compare two runs?
Yes — any two finished runs, not just consecutive ones. Open Compare in the dashboard, pick a before and an after, and you get both scores side by side plus every finding sorted into resolved, still there, and new. Focused re-runs go further and deliver a verdict per finding.
Can my AI assistant or CI pipeline run mocktomer?
Yes — mocktomer ships an MCP server. On Claude, add it as a connector and just sign in — OAuth, no key to paste. Everywhere else (ChatGPT, Cursor, CI pipelines), use a scoped API key you mint in Settings: hashed at rest, optional expiry, a daily run cap, and instant revocation. Setup guides live on the Developers page.
What's the catch with 10 free every month?
There isn't one. No card, no trial clock, and the free allowance refreshes to 10 human shoppers every month. We make money when teams want bigger crews or scheduled monitoring — the free tier is how you find out it's worth it.
Can I run it on client sites? (agencies)
Yes — with each client's permission, add their site and run your whole portfolio from one workspace. The Scale plan's 50 site slots exist exactly for this; need more, just email us.
Can I put my own logo on reports?
On the Scale plan, yes — flip on white-label in Settings, add your brand name and logo, and the PDF reports you hand to clients carry your branding. Built for agencies whose clients should see the agency's name on the deliverable, not ours.
Which AI assistants do the AI shoppers test?
ChatGPT, Claude, and Gemini — through their commercial APIs. Each AI shopper asks a real shopper question ("what's the best one?") and we record whether you get named, where you rank, and who gets recommended instead. AI shoppers come with every paid plan.
How is the AI Readiness Score computed?
Five sub-scores rolled into one 0–100 number: visibility (do assistants recommend you), findability, readability, actionability, and checkout reachability. Every claim in the report links back to the actual probe, so nothing is a black box.
Where do the competitor names in the AI report come from?
Straight from the assistants' own answers — they're whoever ChatGPT, Claude, or Gemini actually recommended when asked a shopper question in your category. We keep no competitor database and add nothing of our own.

The rest is faster to see than to read

10human shoppers free every month — watch them use your site and read why they'd leave, in their own words.

Start free Read a sample report first