Site Audit Pro Fixes: Cross-Page Problems on Larger Sites
Site Audit Pro is built for sites of fifty pages or more, where the hardest problems only show up when you compare pages with each other.
It scores Core Web Vitals for each page rather than the domain as a whole, finds titles, descriptions and body text that are shared between URLs, validates every structured-data block against schema.org, and checks internal links for orphans, broken targets and pages buried too deep. This page explains each of those five finding types in its own section, so older links to them still work, and suggests the order to tackle them in.
Which findings to fix first
On a large site, every Pro finding type can produce a long list. The question is not which list is longest but which problem is costing you the most. Jump straight to a finding:
Problems that decide which page ranks
- Duplicate title tags and meta descriptions: several pages presenting themselves with the same label, so search engines struggle to choose between them.
- Cross-page duplicate content: near-identical body text on different URLs, splitting signals that would be stronger on one page.
Problems that decide whether a page is found
- Internal link problems: orphan pages, links to URLs that no longer exist, and important pages too many clicks from the homepage.
Problems that affect how a page performs and appears
- Per-page Core Web Vitals failures: the specific URLs that are slow to load, slow to respond or visually unstable.
- Invalid structured data: markup that is present but broken, so it cannot support rich results.
As a rule, fix duplication first, because while two pages compete for the same searches, improving either of them is partly wasted. Then fix linking, so that the page you chose to keep is easy to reach. Speed and structured data come next, starting with the pages that bring in the most visitors or revenue.
Fixing per-page Core Web Vitals failures
A domain-wide speed score hides a lot. A site can look acceptable on average while its product pages, which carry large galleries and review widgets, fail badly. Pro reports Largest Contentful Paint (LCP), Cumulative Layout Shift (CLS) and Interaction to Next Paint (INP) for each page, so you can see exactly which URLs pull the average down. INP replaced First Input Delay as a Core Web Vital in March 2024, so older advice that mentions FID is out of date.
- Sort worst first, then filter to the pages that matter commercially. Fix a failing high-traffic page before several failing pages that get little traffic.
- Look for shared causes. If every failing URL uses the same template, the fix belongs in that template.
- LCP: find the largest element above the fold, usually a hero image or headline block. Compress and correctly size the image, avoid lazy-loading it, preload it if it is discovered late, and reduce server response time.
- CLS: give images, videos and embeds width and height attributes, reserve space for adverts, cookie banners and anything injected after load, and avoid fonts that swap in at a very different size.
- INP: reduce the JavaScript that runs when someone taps or clicks. Break long tasks into smaller pieces, defer scripts that are not needed straight away, and review third-party widgets.
- Re-run Pro to confirm the page has moved from failing to passing.
Lab measurements from a tool and field data from real visitors can disagree, because real devices and connections vary. Use Pro to find which pages to work on, and Search Console's Core Web Vitals report to confirm the effect on real users over the following weeks. The Core Web Vitals fixes go into each metric in more depth.
Fixing duplicate title tags and meta descriptions
When several pages share a title, a search engine has no simple way to tell which one should appear for the words in it. It may pick one and ignore the others, or rotate between them, and neither is what you want. Pro compares titles and descriptions across every page it crawled and lists each collision with the URLs involved.
- Take each group of pages sharing a title and ask what makes each one different: a location, a product variant, a service, a stage in a process.
- Rewrite each title to lead with that difference. "Boiler Servicing in Chesterfield" and "Boiler Servicing in Sheffield" are distinct; "Boiler Servicing | Brand" on both is not.
- If you cannot say what makes two pages different, that is a sign they are duplicate content rather than a title problem. Move to the next section.
- Do the same for meta descriptions. They are not used as a ranking signal, but identical descriptions look repetitive in results and do little to persuade anyone to click.
- Check where the duplicates come from. Pagination, sort orders, tag archives and filtered category pages often inherit the parent page's title automatically.
Fixing cross-page duplicate content
Duplicate content is rarely a penalty issue; search engines generally just pick one version to show. The cost is subtler: links, engagement and relevance get split across several URLs instead of building up on one. Pro flags pages whose main content is identical or very close.
For each flagged group, choose one of three remedies:
| Remedy | When to use it | How |
|---|---|---|
| Consolidate | Several thin pages cover the same topic and none needs to exist separately. | Merge the best material into one stronger page and 301-redirect the others to it. Update internal links to point at the survivor. |
| Canonicalise | The duplicates need to exist for visitors, such as printer versions, tracking parameters or sort orders. | Add a rel="canonical" tag on each variant pointing at
the main version. |
| Differentiate | Each page serves a genuinely different need, such as separate towns or product models. | Give each page content that only applies to it: local details, specifications, photos, prices, answers to its own questions. |
Location pages are the most common case on small business sites. Swapping the town name into otherwise identical text is exactly the pattern Pro will flag, and it also reads poorly to a real visitor from that town. Genuine local detail, such as areas covered, travel times, local projects and reviews from customers there, solves both problems.
Remember that a canonical tag is a strong hint, not an order. If the pages are very different, or internal links all point at the non-canonical version, a search engine may choose its own canonical.
Fixing invalid structured data
Structured data can look perfectly fine in a page's source and still fail. A missing comma breaks a whole JSON-LD block; a required property left out, or a date in the wrong format, can stop a page being eligible for the rich result the markup was meant to earn. Pro validates every structured-data block on every crawled page against the schema.org vocabulary, which is how template-wide faults show up.
- Open the Pro report and group errors by type. The same error on every product page means the product template is generating it.
- Fix malformed JSON first: unescaped quotes in product names, trailing commas and unclosed brackets. These invalidate the whole block.
- Then add missing required properties and correct value types, such as numbers stored as text, prices with currency symbols in the number field, or dates that are not in ISO 8601 format.
- Check that the markup matches what is visible on the page. A rating, price or availability in the markup that the visitor cannot see goes against search engine guidelines.
- Test one corrected page in the Schema Debugger before re-running the full Pro audit.
Validity is necessary but not sufficient. Google has cut back several rich result types in recent years: HowTo rich results were removed in 2023, and FAQ rich results are now limited to well-known, authoritative government and health sites. Valid markup for those types still helps machines understand a page, but do not expect it to change how the result looks. The Schema Debugger fixes cover individual schema errors in more detail.
Fixing internal link problems
On a small site, every page tends to be linked from the menu. As a site grows past fifty or a hundred pages, links thin out, and some pages end up reachable only through the sitemap or not at all. Pro checks three things.
Orphan pages
An orphan has no internal links pointing to it. Search engines may still find it through the sitemap, but it receives no internal signals about its importance and visitors cannot reach it by browsing. For each orphan, either link to it from two or three relevant pages, using descriptive anchor text, or, if it no longer serves a purpose, redirect it to the closest current page or remove it.
Broken internal links
A link to a page that returns an error wastes a visitor's click and a crawler's request. Update the link to the correct current URL. If many links point at a page that moved, add a redirect as a safety net, but still correct the links themselves so they do not go through a redirect.
Pages buried too deep
Click depth is the number of clicks from the homepage to reach a page. Important pages several clicks down tend to be crawled less often and treated as less important. Bring them closer through the main navigation, category or hub pages, and contextual links from popular articles.
For a deeper look at the link graph, including anchor text and how authority flows between pages, see the Internal Links fixes.
Order of work
- Run Site Audit Pro across the whole site, ideally before a migration, redesign or quarterly review so you have a baseline.
- Resolve duplicate content groups: decide for each whether to consolidate, canonicalise or differentiate, and put redirects in place.
- Rewrite duplicate titles and descriptions on the pages that remain.
- Fix internal links: point broken links at live URLs, link orphans in, and bring key pages closer to the homepage.
- Work through per-page Core Web Vitals, starting with shared templates and your busiest URLs.
- Correct structured-data errors at the template level and check a sample in the Schema Debugger.
- Re-run Pro and compare with the baseline.
If you have not yet run the standard Site Audit, its single-page checks for titles, headings and alt text are a useful first pass; Pro then adds the comparisons between pages. The HTML Checker Guide helps when a structured-data or heading problem turns out to be caused by broken markup. For the full method behind Pro, read the Site Audit Pro Guide.
1. What is the difference between Site Audit and Site Audit Pro?
The standard Site Audit looks at what is wrong on individual pages, such as a missing title or an image without alt text. Site Audit Pro is aimed at larger sites and looks across pages: it scores Core Web Vitals per URL, finds titles and content shared between pages, validates every structured-data block and checks internal links for orphans, broken targets and deep pages. Many people run the standard audit first and then use Pro for the site-wide view.
2. Does duplicate content get my site penalised?
Ordinary duplication, such as the same product in two categories or a printer-friendly version of a page, does not lead to a penalty. Search engines usually choose one version to show and filter out the rest. The real cost is that links and relevance are split between URLs, and the version chosen may not be the one you wanted. Deliberately copying content from other sites at scale to manipulate rankings is a different matter and can lead to action against the site.
3. Why does my site pass Core Web Vitals overall but Pro shows failing pages?
A site-wide or domain-level figure summarises many pages together, so a large number of fast pages can hide a smaller group of slow ones. Search Console also groups similar URLs, which can blur the picture further. Pro reports each page individually so that the slow ones stand out. Those pages matter if they are the ones people land on, such as product or service pages, even when the overall picture looks healthy.
4. Should I use a canonical tag or a 301 redirect for duplicate pages?
Use a 301 redirect when the duplicate URL does not need to exist for visitors, for example an old page replaced by a better one. The redirect sends people and search engines to the right place and consolidates signals firmly. Use a canonical tag when visitors still need the duplicate URL, such as a filtered or sorted listing or a page with tracking parameters. The canonical tells search engines which version to index while leaving the variant usable.
5. My structured data validates. Why am I not getting rich results?
Valid markup makes a page eligible, but search engines decide whether to show a rich result based on many factors, including the quality of the page, the type of search and their own policies. Some types have also been reduced: HowTo rich results were removed in 2023, and FAQ rich results are limited to well-known authoritative government and health sites. Check that the markup matches visible content and that the page is indexed, then give it time.
6. How many clicks from the homepage should important pages be?
There is no official rule, but a common working target is that key pages such as main services, top categories and best-selling products are reachable within about three clicks. The idea is less about a precise number and more about making sure important pages are linked from places that are themselves well linked. Hub pages, clear category navigation and contextual links from popular articles are the usual ways to bring deep pages closer to the surface.
7. How often should I run Site Audit Pro?
A quarterly run suits most growing sites, with extra runs before and after a migration, redesign, change of platform or large content project. Running it before a change gives you a baseline, and running it afterwards shows whether the change created new duplicates, broke internal links or slowed down key templates. Sites that add many pages every week, such as large shops or publishers, may find a monthly check catches problems while they are still small.
Site Audit Pro crawls your whole site to score Core Web Vitals per page, find duplicate titles and content, validate every structured-data block and check internal links. Pay as you go, no subscription.
Run the Site Audit Pro →