Issue reference

How to fix: Restore the page, or 301-redirect it to the closest live equivalent. Then update the internal links listed under Inlinks so they point at the live URL directly.

How to fix: Check server and application logs for the URL at the time of the crawl. If the error was temporary, lower crawl speed and re-crawl; if it repeats, fix the failing route before search engines see it.

How to fix: Confirm the host resolves and accepts connections from outside your network. Look for firewall or bot-protection rules blocking the crawler, and raise the request timeout if the page is simply slow.

How to fix: Replace the link with a working URL, link to an archived copy, or remove it. Outbound links to dead pages look neglected to visitors.

How to fix: Open the link yourself. If the other site is down for good, replace or remove the link; if it was a blip, re-check it in your next crawl.

How to fix: Check whether the domain still exists. Expired or parked domains should be unlinked; slow sites may only need a longer request timeout.

How to fix: If the URL should be crawled and indexed, remove or narrow the Disallow rule that matches it. If it is meant to be private, consider noindex instead, since blocked URLs can still be indexed from links.

How to fix: Profile the page's server time: slow database queries, uncached rendering and heavy middleware are the usual causes. Add caching for pages that change rarely.

How to fix: Update internal links to point straight at the final URL. The redirect can stay for visitors arriving from old bookmarks and external links.

How to fix: Update the link to the URL it redirects to. External redirects often mean the other site moved the page, so check it still says what you link it for.

How to fix: Point the first URL directly at the final destination with a single 301, and update internal links to the final URL.

How to fix: Trace the chain and remove the rule that sends a URL back to an earlier step. Loops often come from conflicting trailing-slash, www or HTTPS rules.

How to fix: Add a unique <title> in the page <head> that describes the page in plain words, ideally with the main topic first.

How to fix: Keep one <title> element. Duplicates usually come from a theme and a plugin both writing one; switch one of them off.

How to fix: Write a distinct title for each page. For paginated or filtered pages, add the page number or filter, or canonicalise them to the main page.

How to fix: Shorten the title so the important words come first and it fits in search results. Drop repeated brand names and filler.

How to fix: Add useful detail such as the product type, location or brand so the title explains what the page is.

How to fix: Usually fine. If you want more room in search results, make the title a slightly fuller version of the H1, for example adding the brand or category.

How to fix: Add a meta description that summarises the page and gives a reason to click. Search engines may still rewrite it, but a good one is often used.

How to fix: Keep one meta description tag. Check templates and SEO plugins for a second one being injected.

How to fix: Write a description specific to each page. For large templated sections, generate it from page data such as product name and price.

How to fix: Trim the description so the key message sits in the first sentence and is not cut off.

How to fix: Expand it to a full sentence or two that tells searchers what they will find on the page.

How to fix: Add one visible H1 that names the page's main topic. Check that the template does not render the heading as a styled <div>.

How to fix: Use one H1 for the page title and H2 to H6 for sections below it. Logos and sidebars marked up as H1 are a common cause.

How to fix: Give each page a heading that reflects its own content, especially across products, locations and paginated lists.

How to fix: Shorten the heading to the page's core topic and move supporting detail into an intro paragraph.

How to fix: If the page should appear in search, remove noindex from the meta robots tag or the X-Robots-Tag header. If it is meant to be hidden, no action is needed.

How to fix: Remove page-level nofollow unless you deliberately want search engines to ignore every link on the page. Use rel attributes on individual links instead.

How to fix: Decide whether the page should be indexed. If not, reduce internal links to it so crawl effort goes to pages you want found; if so, remove noindex.

How to fix: Add a self-referencing <link rel="canonical"> with the page's preferred absolute URL. It protects against duplicates created by tracking parameters.

How to fix: Output a single canonical, in the <head> or the HTTP header but not both with different values. Conflicting canonicals are usually ignored.

How to fix: Expected for parameter, print or variant URLs. If this page should rank on its own, change its canonical to point at itself.

How to fix: Point the canonical at a live URL that returns 200. If the target moved, use its final address rather than a redirecting or missing one.

How to fix: Add the matching hreflang annotation on the other page so each alternate links back. Hreflang only counts when both pages confirm the pair.

How to fix: Use an ISO 639-1 language code, optionally followed by an ISO 3166-1 alpha-2 region (for example en or en-gb), or x-default. Regions alone, such as uk, are invalid.

How to fix: Update the annotation to the alternate's final, live URL. Remove it if that language version no longer exists.

How to fix: Include the page itself in its own hreflang set, using its own language code and URL.

How to fix: Add an x-default annotation pointing at your language selector or main fallback page for visitors whose language you do not target.

How to fix: Make the <html lang> attribute and the page's self-referencing hreflang code agree, and check both match the language the content is written in.

How to fix: Choose one version to keep. Redirect or canonicalise the others to it, or rewrite them so each page has its own content.

How to fix: Add substantive content that answers what visitors came for, merge the page with a related one, or noindex it if it exists only for navigation.

How to fix: Compare the pages listed in the near-duplicate group. Merge them, canonicalise the weaker versions to the strongest, or rewrite the shared boilerplate so each page says something of its own.

How to fix: Add alt text that describes each meaningful image. Use an empty alt="" for purely decorative images so screen readers skip them.

How to fix: Run the block named in the details through a JSON validator and fix the syntax, usually a trailing comma, an unescaped quote or a template that printed nothing. For Microdata and RDFa, check itemscope/itemprop or typeof/property nesting.

How to fix: Add the required property listed in the details, or correct its value type (for example a number instead of text, or a full URL). Items with errors are not eligible for that rich result.

How to fix: Add the recommended property if you have the data. The item still qualifies without it, but richer results usually need it.

How to fix: Check the spelling and capitalisation against schema.org (types are PascalCase, properties camelCase). Remove vocabulary that search engines do not understand.

How to fix: Fix each validation error listed in the details, most often disallowed scripts, inline styles over the size limit or missing required tags. Invalid AMP pages are not served from AMP caches.

How to fix: Add <link rel="canonical"> on the AMP page pointing at the regular page (or at itself if the AMP page is the only version), and rel=amphtml on the regular page pointing back.

How to fix: Fix these first: they block some visitors completely, for example form fields with no label or controls that cannot be reached by keyboard. The details list each axe-core rule with the affected elements.

How to fix: Work through the rules in the details, typically low colour contrast, images without text alternatives or missing page language. Fix them in shared templates to clear many pages at once.

How to fix: Address these after critical and serious issues: landmarks, heading order and duplicate ids usually make up this group and are cheap to fix in templates.

How to fix: Low-impact refinements. Fix them when you next touch the template, and re-run the audit to confirm.

How to fix: Add <meta name="viewport" content="width=device-width, initial-scale=1"> to the page <head>, and do not disable zoom.

How to fix: Find the element that overflows: wide tables, fixed-width containers and large images are typical. Use max-width: 100%, responsive tables or horizontal scrolling inside the component rather than the page.

How to fix: Open the Lighthouse audit named in the details. Common fixes are larger tap targets with more spacing, legible font sizes of at least 12 px and content that fits the viewport.

How to fix: Start with the biggest opportunities Lighthouse lists: compress and resize images, remove unused JavaScript and CSS, and defer third-party scripts. Re-test on the same device setting to compare.

How to fix: Make the largest element load sooner: serve the hero image in a modern format at the right size, preload it, avoid lazy-loading it, and cut server response time and render-blocking CSS.

How to fix: Break up long JavaScript tasks, defer non-essential scripts and avoid heavy work in input handlers so the page responds to taps and clicks within a frame or two.

How to fix: Reserve space for images, ads and embeds with width and height or aspect-ratio, avoid inserting content above existing content, and use font-display with matched fallback fonts.

How to fix: Review the words listed in the details and correct genuine mistakes in the page copy or its template. Brand names and jargon may be false positives; check the page language is set correctly.

How to fix: Review the flagged sentences. Fix real errors; style suggestions are optional and can be ignored where the wording is intentional.

How to fix: Read the coverage state in the details. Fix any blocker it names (noindex, robots.txt, redirect, soft 404), make sure the page is linked internally and in your sitemap, then request indexing in Search Console.

How to fix: Google picked another URL as canonical. Make the signals agree: consistent internal links, the sitemap and rel=canonical should all point at the URL you want indexed, and near-identical pages should be consolidated.

How to fix: The page is indexable but did not appear in search during the period. Check it is indexed, targets a query people search for and is linked from relevant pages; otherwise consider merging or noindexing it.