Free tool, no sign-up
Free SEO checker for any website
Enter any public web address. This reads the site’s sitemap, checks up to twenty pages the way a crawler does, and reports what it actually finds: titles and descriptions, heading structure, missing alt text, canonical and indexing directives, robots.txt and the sitemap, and Google’s own performance scores. Problems are grouped by fault rather than by page, so the same issue on twelve pages is one line rather than twelve.
No account. No email address. Nothing is stored.
What the report covers
Four groups of checks, across every page it reads
Every check comes back as a pass, a warning or a failure, with a sentence saying what it means rather than a code you have to look up. The on-page and technical checks run on every page found. The performance numbers take longer, because Google is loading the page in a real browser to produce them, so they fill in on their own once they arrive.
There is also a single-page option if you only want one address looked at, which is faster and checks the links on that page alone.
Reading the whole site also finds things a single page cannot show you. Duplicate title tags and duplicate meta descriptions only exist as a comparison between pages, and they are among the most common real faults on a WordPress site: archive pages, paginated lists and near-identical service pages all produce them, and pages sharing a title compete with each other in search results. The report also lists the slowest pages of those it read, and a tally of every check that ran and passed.
On-page
The title tag and meta description, with the text that was actually found and whether the length works in a search result. Whether the page has exactly one H1 or none. How many images are missing an alt attribute, with examples. How much visible text the page carries, which catches a page that published empty.
Technical
The canonical tag and where it points. Robots directives in both the meta tag and the HTTP header, because a noindex in either one removes the page from search entirely. Whether structured data is present. Whether every image, script and stylesheet loads over https, since one http resource on a secure page costs the padlock. Then robots.txt and the XML sitemap, fetched from the site itself.
Performance
Google’s own Performance, SEO and Accessibility scores for mobile and desktop, from the same engine that powers PageSpeed Insights. Underneath them, the lab timings and, where Google holds it, what real Chrome users measured over the last month.
Links
Internal links from every page read are pooled together, de-duplicated, and a sample is requested one at a time. Anything answering 404 or 500 is listed with its status, along with any page in the sitemap that would not load. Links pointing at pages already fetched are not tested twice.
Reading the performance numbers
A lab score is not what your visitors experienced
Two different kinds of number come back from Google, and treating them as the same thing is the most common mistake made with a speed report. This tool labels them separately for that reason.
Lab measurements come from loading the page once in a simulated browser on a throttled connection. They always exist, for any page, and they are reproducible enough to test a fix against. They are also a single synthetic load from one location, so a lab score of 40 does not mean your visitors wait as long as it suggests, and a lab score of 95 does not prove they do not.
Real visitor data comes from Chrome users who actually loaded your page over the previous 28 days. This is the data Google uses when it assesses Core Web Vitals, so it is the one that carries weight. It only exists for sites with enough traffic to make the sample meaningful, which in practice means roughly a thousand visits a month. Below that, the section will tell you Google holds no data. That is normal and is not a fault with the site.
One metric appears only under real visitor data: Interaction to Next Paint, which replaced First Input Delay as a Core Web Vital. It measures how long the page takes to respond after somebody clicks or taps, and it cannot be measured in a lab at all, because a lab run has nobody in it to do the clicking.
Being straight about it
What this tool cannot tell you
A free tool that claims to audit your whole search presence in ten seconds is not being honest with you. Here is what this one genuinely does not do.
- It checks up to twenty pages, not necessarily all of them. On a larger site the report says how many pages exist and how many were read. A fault on page fifty will not appear.
- The link check is a sample, not a full audit. Internal links are pooled from the pages read and up to thirty are tested. A site with two thousand links is not exhaustively checked, and external links are not tested at all.
- Performance is measured on one page only. Google measures a single URL at a time, so the scores describe the address you entered rather than the whole site. Another page can be much slower.
- It has no ranking data. It cannot tell you what position you hold for anything, because that requires querying search results at scale from many locations. Search Console is where that answer lives, and it is free.
- It has no backlink data. Who links to you, and how much that is worth, comes from commercial crawl datasets that cost real money. Anything free claiming to have it is showing you a fragment.
- It has no keyword or competitor data. Search volumes and difficulty scores are licensed data, not something that can be read off your page.
- It confirms structured data exists, it does not validate it. Whether your markup is correct is a question for Google’s Rich Results Test, which is the authority on it.
- It cannot judge whether your content is any good. No automated check can. A page can pass every test on this list and still deserve to rank nowhere, because nobody needed it written.
One more thing worth saying plainly, because this page will find people searching for SEO help. This studio builds and repairs WordPress. It does not sell SEO campaigns, content writing or link building, and it will not quote for them. What it fixes is the engineering underneath: the reasons a page loads slowly, returns the wrong status, blocks its own indexing, or breaks when the traffic arrives.
After the report
What to do about what it found
Most of what this tool flags is fixable in an afternoon by whoever maintains the site. A missing meta description, an empty alt attribute, a sitemap that was never declared in robots.txt: none of that needs an agency. Three findings are worth more attention.
A noindex you did not put there
If the indexing check fails, nothing else on the report matters until it is resolved. A noindex directive removes the page from search completely, and it is most often left behind by a staging site that was copied to production with the discourage-search-engines setting still on. It is a five-minute fix and it is the single most expensive thing on this list to leave alone.
A performance score in the red
A poor mobile score is rarely solved by installing another caching plugin. It is usually a specific bottleneck: uncompressed images served at full size, a page builder loading its entire library on every page, render-blocking scripts in the head, or a database doing hundreds of queries to build one page. Finding which one it is takes profiling rather than guessing. That is WordPress speed optimization, and it starts at $349.
Broken internal links, or a page that would not load at all
Broken internal links waste crawl budget and dead-end visitors, and a handful of them on one page usually means many more across the site. If the tool could not load your page at all, or the site is behaving in ways nobody can explain, that is worth looking at properly. Unexpected redirects, pages that appear and disappear, and content nobody remembers publishing are all early signs of a compromise, which is malware removal territory from $199.
If you would rather someone read the whole thing properly, the free written audit covers the site rather than one page, and comes back with one fixed price. Roughly one audit in five ends with us saying the work is not worth paying for.
Questions
Common questions about the checker
Is it really free, and is there a limit?
Free, with no account and no email address. There is a limit of ten checks an hour from the same connection, which exists so the tool cannot be used as a free bulk scanning service at our expense. A whole-site check counts as one, however many pages it reads. If you hit the limit, the message tells you how long is left.
Does it check my whole website, or just one page?
The whole site, up to twenty pages. Enter one address and it reads that site’s XML sitemap, works out which pages exist, and checks each one. You do not enter pages one at a time. If there is no usable sitemap it follows links from the address you gave instead.
Twenty is a cap, not a target. A five-page site gets all five. A two-hundred-page site gets the first twenty and the report says so. The cap exists because every page is a request made by our server on your behalf, and an uncapped tool would be a free scanning service for anyone who wanted one. There is also a single-page option if that is all you need.
Why is the speed score only for one page?
Because Google measures one URL at a time. There is no such thing as a speed score for a whole website. Running it on twenty pages would mean forty separate measurements and about fifteen minutes of waiting, so the score is for the address you entered. If you want a different page measured, enter that page.
Does it only work on WordPress sites?
No. Every check reads the HTML that the address returns, so it works on any public website whatever it is built with. The advice attached to a failing check is written by people who fix WordPress, so that part is most useful if that is what you are running.
Do you store the addresses I check, or the reports?
No. There is no database table behind this tool and reports are not saved anywhere. The address you submit is sent to Google’s PageSpeed Insights API to produce the performance section, which is unavoidable, and a counter for the hourly limit is held against a one-way hash of your IP address for one hour. Nothing else is kept, and there is nothing to email you afterwards because we do not have an address for you.
Can I check a site that is not mine?
Yes. It only reads pages that are already public, in exactly the way any search engine does. It does not attempt to log in, does not submit forms, and does not touch anything a visitor could not reach.
Why did it say my page could not be loaded?
Usually one of three things. The site took more than fifteen seconds to answer, which is itself worth knowing. Or it is behind a firewall that blocks automated requests, which many hosts turn on by default. Or the address returned an error status, in which case the report tells you which one. A password-protected or staging site cannot be checked, because the tool sees only what an anonymous visitor sees.
Why is the performance section slower than the rest?
Because Google is genuinely loading your page in a browser to measure it, twice, once as a phone and once as a desktop. That takes fifteen to forty seconds each and there is no way to hurry it. The rest of the report does not wait for it, which is why the page fills in the fast checks first and the scores drop in afterwards.
My title is 70 characters and it says that is too long. Does it matter?
Only a little. Google rewrites titles in results a large share of the time regardless of length, and a long title is not a penalty. The check exists because a title cut off mid-sentence in a result reads badly to a human deciding whether to click, and that is the actual cost. Treat it as a presentation note rather than a fault.
Can I use this tool on my own site?
It is a WordPress plugin, and it is not currently distributed. If you maintain sites and would find it useful, say so in a message and we will tell you where it stands.
The report found something. Now what?
Send us the address and we will write a free audit of the whole site, not one page: every problem we found ordered by what it costs you, what we would do about each one, the warnings you can safely ignore, and one fixed price in US dollars agreed before anything starts. It comes back as a document. There is no sales call.