Every check Fullsweep runs on your site

Address moves visitors to a different page

This address does not serve a page of its own. It sends everyone who asks for it, including Google, on to a different address.

Page is missing

The site was asked for this address and said the page is not there. Anyone who follows a link to it, including Google, hits a dead end.

Page failed to load

The site tried to build this page and failed. Search engines drop pages that keep failing, and visitors see an error instead of your content.

Page took too long to answer

This page did not answer before we gave up waiting. Google gives up too, so a page this slow may never make it into search results.

Redirect passes through more than one stop

This address sends visitors to a second address, which sends them on again. Every extra stop costs load time, and Google may stop following after a few hops.

Redirect goes in circles

This address redirects back to an address it has already passed through, so it never reaches a page. Nobody can open it and Google cannot index anything here.

Page is served over an insecure address

This page loads over http rather than https. Browsers label it as not secure, and Google prefers the secure version of any page.

Web address contains a space

This address has a space in it, either written out or as %20. Spaces break links when people copy, share or paste them into other tools.

Web address has a double slash

The path in this address has two slashes in a row, such as /blog//post. Most sites serve the same page at both spellings, which splits one page into two in Google's eyes.

Page has no title

This page has no title tag, so Google writes its own from whatever text it can find. You lose control of the one line most people read before deciding to click.

Title is under 20 letters

This page's title is shorter than 20 letters, so it wastes the space you get in search results. It usually leaves out the words people actually search for.

Title is over 60 letters

This page's title runs past 60 letters, so Google will probably cut the end off. Whatever you put last may never be seen.

Page has more than one title

This page carries two or more title tags. Google picks one and ignores the rest, and it may not pick the one you wrote.

Pages share the same title

These pages use exactly the same title. Google has to guess which one to show for a search, and it often shows none of them well.

Page has no meta description

This page has no meta description. Google fills the gap with a sentence lifted from the page, which often reads badly and rarely earns the click.

Meta description is under 80 letters

This meta description is shorter than 80 letters, so most of the search result sits empty. The page then sells itself less well than the ones beside it.

Meta description is over 160 letters

This description runs past 160 letters, so Google is likely to cut it short. The last part, which is often the reason to click, disappears.

Page has more than one meta description

This page carries two or more meta descriptions. Google picks one, and it may well pick the one you did not write.

Pages share the same meta description

These pages use the same meta description word for word. It tells a searcher nothing, and wastes the one line you control in the result.

Page has more than one canonical tag

This page names more than one canonical address. When canonical tags disagree, Google usually ignores all of them and decides for itself.

Canonical tag uses a shortened address

The canonical tag here gives a partial address instead of the full one. Different tools fill in the missing part differently, so the tag can end up naming a page you never meant.

Two canonical tags disagree with each other

This page names one canonical address in its code and a different one in the settings sent by the site. Google sees a contradiction and may ignore both.

Canonical points at a page that points somewhere else

This page names another page as the canonical one, but that page names a third. Google has to follow a trail to work out what to keep, and it often abandons the trail.

Canonical points at a page that redirects

This page names a canonical address that does not serve a page. That address redirects somewhere else, so Google ends up at a different page than the one you named.

Canonical points at a broken page

This page names a canonical address that is missing or failing. Google is being told that the important version of this page is one that does not load.

Page settings sit too far down the code

This page has meta tags in the body instead of the head. Crawlers ignore meta tags found down there, so whatever those tags say is having no effect at all.

Sitemap lists a page set to noindex

Your sitemap points search engines at this page, while the page itself asks to be kept out of results. The two instructions contradict each other, and the sitemap spends attention on a page that can never rank.

Sitemap lists a page blocked in robots.txt

Your sitemap points search engines at this address, and your robots.txt tells them not to read it. They cannot open the page, so the listing does nothing except weaken the rest of the sitemap.

Sitemap lists a page that names another as canonical

This address is in your sitemap, but the page itself names a different address as the one that should rank. A sitemap should only list the pages you want in results, so this sends Google to a copy.

Page has no main heading

This page has no H1 heading with any text in it. The H1 is the clearest statement of what a page is about, for readers and for search engines.

Headings skip a level

The headings on this page jump down more than one level at a time, such as H2 straight to H4. That breaks the outline of the page for screen readers and makes the structure harder to follow.

Pages have identical text

These pages carry exactly the same visible text. Search engines pick one to show and largely ignore the others, so the work spent on the rest earns nothing.

Language value is not a real language code

This page states a language, but the value is not a valid code. Anything reading that value, including translation tools and screen readers, will ignore it.

Language set has no fallback

This page lists other language versions but names no x-default. Visitors whose language is not on your list have no page marked out for them.

Language set does not include this page

This page lists other language versions but never lists itself. An hreflang set must name every page in it, including the one carrying the list, or search engines discard it.

Language tags on the page disagree

This page's own hreflang entry names a different language from the page's lang value. One of the two is wrong, and search engines may show this page in the wrong country.

Language link uses a shortened address

One of this page's hreflang entries gives a partial address instead of a full one. Search engines require full addresses here, so that entry is ignored.

Language link is not returned by the other page

This page points at another language version, but that page does not point back. hreflang only counts when both pages name each other, so search engines drop the pairing.

Language link points at a page that points elsewhere

This page names a language version whose canonical tag names a different address. hreflang and canonical are giving search engines conflicting instructions, so the pairing is dropped.

Language link points at a page that redirects

This page names a language version whose address redirects somewhere else. Search engines expect the exact final address here, so the pairing is ignored.

Language link points at a broken page

This page names a language version that is missing or failing. Anyone sent there hits a dead end, and the whole language set is weakened.

One page is claimed as two different languages

Other pages point at this address with hreflang, but they do not agree on what language it is in. Search engines cannot settle the conflict and usually ignore the set.

Page links to insecure addresses

This page has links using http instead of https. They can be tampered with while they travel, and browsers may warn anyone who follows them.

Page links to a test address

This page links to localhost or 127.0.0.1, an address that only works on the machine the page was built on. For every real visitor that link is broken.

Next or previous page is named but not linked

This page names a next or previous page in its code, but offers no link a person could click. Search engines follow real links to find the rest of a series, so the later pages may never be found.

Every link to this page is marked nofollow

This page is meant to be in search results, yet every link to it from your site is nofollow. No ranking value reaches it, so it competes far below its weight.

Links to this page disagree with each other

Some links to this page are marked nofollow and others are not. The instruction is inconsistent, which usually means one group of links was marked by mistake.

Images have no alt text

One or more images on this page have no alt text. People using a screen reader hear nothing about them, and search engines have no idea what they show.

Alt text is over 100 letters

An image here has alt text longer than 100 letters, which a screen reader reads out in full. Alt text that long is usually a caption or a list of keywords, not a description.

Image will not show in some browsers

This page has a picture block with no plain image inside it as a fallback. Anything that does not read the block, including some crawlers, shows nothing at all.

Page is slow to start answering

This page took more than 0.8 seconds just to start answering, before any content had loaded. That delay is added to every visit and to every crawl.

Form sits on an insecure page

This page has a form on it and the page itself loads over http. Browsers show a not secure warning on forms like this, and many people will refuse to fill one in.

Form sends answers to an insecure address

A form on this page sends what people type to an http address. Anything typed in, including names, emails and passwords, can be read while it travels.

Web address uses underscores

This address contains an underscore, which search engines do not read as a space between words. The words in the address therefore run together.

Canonical points at a page search engines are told to skip

This page names a canonical page that is set to noindex or is blocked in robots.txt. You are pointing Google at a page it has been told to leave out, so neither page is handled properly.

Page tells search engines not to list it

This page carries a noindex instruction, so search engines will keep it out of their results. It cannot bring in any search traffic while that is true.

Page is blocked in robots.txt

Your robots.txt file asks search engines to stay away from this address. They cannot read the page, so it will not rank for anything written on it.

Page blocks its own search preview

This page carries a nosnippet instruction. Google can still list it, but cannot show any preview text under the link, which usually cuts clicks sharply.

Page blocks its images from image search

This page tells search engines not to list its images. Any visitors those images could bring in from image search are turned away.

Page has under 200 words

The visible text on this page comes to fewer than 200 words. Thin pages rarely rank, because there is little for a search engine to match a search against.

Page is built from too many parts

The code behind this page contains more than 1,500 separate parts. Very heavy pages load slowly and take search engines longer to read.

The same name is used twice on the page

Two or more parts of this page share the same id, the unique name given to one part. Links that jump to a section, and anything else relying on that name, will land on the wrong one.

Page does not say what language it is in

This page does not state a language, so screen readers can read it in the wrong accent. Browsers may also offer to translate a page the reader already understands.

Language link points at a page search engines are told to skip

This page points at a language version that is set to noindex or blocked in robots.txt. The pairing cannot work, because that page is not allowed in search results at all.

Page links to more than 100 of its own pages

This page carries more than 100 links to other pages on the same site. The more links a page has, the less weight each one passes on, and the harder it is to use.

Page tells search engines not to follow its own links

This page marks one or more links to its own site as nofollow. That stops this page's ranking value passing on to the pages it links to.

Links to other sites are not marked nofollow

This page links out to other sites without a nofollow mark. Some of this page's ranking value flows out to them.

Page links nowhere

This page has no links leaving it at all, so anyone landing here has nowhere to go next. None of its ranking value passes on to the rest of the site.

No page on the site links here

Nothing on the site links to this page, so a visitor can only reach it by knowing the address. Pages nobody links to are found late, crawled rarely and rank poorly.

Link to another site passes through a redirect

A link on this page points at an address that sends visitors on somewhere else before they arrive. Every extra stop costs load time, and the site you meant to link to may not be where the link now ends up.

Link to another site is broken

A link on this page points at a page on another site that is no longer there. Either that page is gone, or the whole address stopped working. Readers hit a dead end, and a page carrying dead links looks neglected to people and to search engines.

Images do not state their size

One or more images here do not say how wide and tall they are. The page jumps about while they load, which annoys readers and counts against the page in Google's page experience measures.

Image on the page is over 500KB

One of the images on this page weighs more than 500KB. Heavy pictures are the most common reason a page feels slow, and the delay is worst on a phone, which is the version Google judges you on.

Page takes more than four clicks to reach

Getting from the home page to this one takes more than four clicks. Pages this deep are crawled less often and receive less of the site's ranking value.

Site does not insist on the secure version

This page loads securely, but the site never tells browsers to insist on the secure version. Until they are told, a browser can still try the insecure address first.

Site does not limit what can run on the page

This page does not send a Content-Security-Policy, the setting that lists which outside code a page is allowed to load. Without it, anything injected by an attacker or a hacked plugin runs freely.

Site lets browsers guess what a file is

This page does not send the X-Content-Type-Options setting, so a browser may guess at a file's type. It can then treat an uploaded file as code, which is a known way in for attackers.

File type does not match the address

This address ends in one file type but the site calls it another, such as .html sent as plain text. Search engines may then refuse to treat it as a page.

Page is not set up for phones

This page has no viewport tag, so phones show it zoomed out at desktop width. Google looks at the phone version of your site first, so this affects every ranking the page has.