Questions and answers

No. Reports run without signing up. An account is only needed to keep history, compare a page over time or raise the daily limit.

Usually under a minute. Most of that is the Core Web Vitals lookup, which queries Google's own field data and cannot be hurried. A page analysed recently returns immediately from cache.

Any page reachable over the public internet. Pages behind a login, on a private network or blocked by robots.txt cannot be fetched, the report will say so rather than guess.

The markup checks read the HTML as served, which is what a crawler gets first. Core Web Vitals and the page capture come from a full browser render so both views are represented.

Five weighted categories: crawling and indexing, content and metadata, Core Web Vitals, schema and links and media. Every check that contributes is listed in the report so the number can always be traced back to the findings behind it.

Google truncates titles on rendered width, not character count. A 60-character title of wide letters can be cut where a 70-character one is not so the report measures the pixels and shows where the cut falls.

Lab data is one synthetic load on an emulated phone. Real-user data is what Chrome recorded from people who actually visited and it only exists once a page has enough traffic for Google to report it. Where both are available the report says which it is showing, because a lab number should never be presented as real experience.

The check had nothing to examine, hreflang on a single-language page, for instance. That is neither a pass nor a failure so it is counted separately rather than inflating the score.

A report checks the page you gave it. A crawl starts at a URL, follows the links it finds on your own domain, and checks every page it reaches. It is how you find the problems on the pages you did not think to look at: the broken link four levels down, the section that lost its canonical tags in a template change.

Up to 100 on Pro and 500 on Agency, per crawl. If a site is larger than that the crawl stops at the limit and tells you, rather than trailing off and leaving you to wonder whether it finished.

No. It reads your robots.txt first and obeys it, including a Crawl-delay if you set one, and it leaves a gap between requests either way. One page is fetched at a time per site. A crawl is slower than it could be on purpose: a tool that knocks your site over while measuring it has failed at the only job it had.

No. Links to other sites are recorded, and checked for whether they still resolve, but they are never crawled. The crawl stays on the host you started it on.

Every five minutes by default, around the clock, from our end. Pro watches up to 3 sites and Agency up to 10.

That is what the confirmation step is for. A single failed check says as much about the network between us as about your server, so nothing is called an outage until a second check agrees. What reaches you is an incident with a start, a duration and the response times either side of it.

Yes, and another when it comes back. The second matters as much as the first: an alert with no all-clear leaves you refreshing a dashboard to find out whether it is over.

Yes, daily or weekly. The point is catching a regression on its own schedule rather than the next time somebody happens to look, which in practice is after a client mentions it.

Everything the dashboard does: run a report, read the findings, start a crawl, add a monitor, read your account. It authenticates with a bearer token you create in the dashboard, and the documentation shows real request and response bodies in six languages rather than a schema dump. Pro includes 1,000 calls a month and Agency 4,000.

Calls are counted separately from reports, but a call that runs a new analysis spends a report as well, because it is one. Reading a report you already have costs a call and nothing else.

The report and the page's HTML as fetched, so a report can be reopened without crawling the page again.

Anyone you send the link to. The address carries a random key and no sign-in, so it works like any share link. Reports are not listed anywhere, and a report on a page inside a site carries a noindex tag, so a URL you have not published stays out of search results. A report on a homepage, which is public by definition, can be indexed. Your history and tracked pages are separate, and need you to be signed in.

One analysis of one URL. Re-running the same URL later counts again, since it fetches the page fresh. Opening a report you already have is free, however often.

No, and that is deliberate rather than an oversight. You buy a term, it runs, and it ends: nothing is charged to you again without you choosing it. We email you three days before a term ends and again on the day, so it lapses because you decided to let it rather than because you forgot.

The account returns to the Free plan. Nothing is deleted: your reports, history and settings stay where they are, and buying another term puts the allowances back. What stops is the paid capability, so saved history, exports and any workspaces over the free limit become read-only.

There is nothing to cancel. Since nothing renews itself, not buying another term is cancelling, and you keep what you paid for until the day it runs out.