SEO audits can find hundreds of issues. Not all of them matter.
The goal is not to fix every issue in this library. The goal is to correctly diagnose the ones affecting your business.
Choose a diagnostic area or use the filters to narrow the full library.
A missing title on a core service page can affect how that page is understood and displayed in search.
The same missing title on an intentionally noindexed utility URL may deserve no attention.
A 404 receiving valuable backlinks can be an authority-recovery opportunity.
A 404 for a URL that should no longer exist may be working exactly as intended.
A duplicate page may need consolidation.
Or it may be a legitimate regional version handled correctly through canonicalization and hreflang.
This SEO Audit Issue Library is designed to answer four questions:
What does the issue mean?
How serious can it be?
How do you confirm that it is actually a problem?
What should you investigate next?
The library covers issues across:
Technical SEO
Crawling
Indexing
Canonicalization
Redirects
Site architecture
Internal linking
Content quality
Search intent
On-page SEO
JavaScript rendering
Core Web Vitals
Mobile SEO
Images and video
Structured data
International SEO
Ecommerce SEO
Local SEO
Google Business Profile
Backlinks
Outbound links
Spam and manual actions
Website security
Analytics and attribution
Site migrations
AI search crawler access
SEO tools detect conditions.
Strategists determine consequences.
A crawler might report:
1,400 missing meta descriptions.
That sounds serious.
But suppose 1,350 belong to non-indexable filter URLs and the remaining 50 belong to product pages with strong search snippets generated from their visible content.
The number 1,400 does not tell you the priority.
Likewise:
10 orphan pages may represent abandoned campaign URLs.
Or one of them may be your highest-margin commercial service.
The second situation matters far more.
Google's own Search Console documentation and recommendations make the same general distinction: not every excluded URL or technical recommendation needs corrective action.
The first question is always:
Should this URL or asset be doing something different from what it is doing now?
This library uses five practical classifications.
| Severity | Meaning |
|---|---|
| Critical | Can remove important pages or major sections from search, create severe security/spam exposure, or materially threaten existing organic revenue |
| High | Can significantly suppress discovery, indexing, rankings, authority, local visibility, conversions, or measurement |
| Medium | Can limit performance but is usually page-, template-, or context-dependent |
| Low | Usually a hygiene, presentation, efficiency, or incremental optimization issue |
| Context | Not inherently wrong. Investigate the page's purpose before changing anything |
Severity is a starting point.
It is not the priority score.
A medium issue on a page responsible for 30% of organic revenue can deserve action before a high-severity condition on irrelevant URLs.
Different problems require different evidence.
A complete audit should not depend on one crawler.
| Evidence source | Best used for |
|---|---|
| Site crawler | Status codes, metadata, canonicals, directives, links, headings, structured data, templates |
| Google Search Console | Indexing, canonical selection, search performance, Core Web Vitals, manual actions, security issues |
| Server logs | Real crawler behavior, crawl frequency, wasted crawling, response failures |
| Rendered crawl/browser inspection | JavaScript content, dynamic links, metadata changes, rendering differences |
| PageSpeed Insights / CrUX | Real-user performance and Core Web Vitals |
| Analytics | Traffic, engagement, conversions, landing-page behavior |
| CRM / revenue data | Lead quality, opportunities, pipeline, revenue |
| Backlink index + Search Console | Link acquisition, lost links, broken backlinks, suspicious link patterns |
| Google Business Profile | Local identity, category, location, reviews, local visibility |
| Manual SERP review | Search intent, competing page types, title rewrites, cannibalization, local results |
| Content review | Accuracy, usefulness, evidence, differentiation, intent alignment |
| Structured-data validators | Syntax, entity implementation, rich-result eligibility |
One crawler can tell you a title is duplicated.
It cannot tell you whether the two pages should exist.
That requires architecture analysis.
Try a shorter search, choose another category, or use a broader term.
Crawling problems affect whether search engines can request pages, resources, and links.
Crawl budget itself should not be treated as an emergency for every site. Google's current crawl-budget guidance is aimed mainly at very large or rapidly changing sites, including sites with roughly 1 million+ pages, fast-changing sites around 10,000+ URLs, or sites with large numbers of discovered-but-not-indexed URLs.
| Issue | What it means | Typical severity | How to diagnose |
|---|---|---|---|
| Important URL blocked by robots.txt | Crawlers are instructed not to request a page that should be discoverable | High | Test the URL in robots.txt and Search Console URL Inspection |
| robots.txt unavailable | Crawlers cannot reliably retrieve crawling instructions | High | Request /robots.txt, inspect server response and crawl logs |
| robots.txt returns 5xx | Server failure prevents reliable processing of crawl rules | High | Check HTTP response and server logs |
| Invalid robots.txt rule | Directive is malformed, unsupported, or does not match intended URLs | Medium | Test rules against example URLs |
| Important CSS or JS blocked | Rendering resources needed to understand a page cannot be requested | Medium–High | Compare rendered page and blocked-resource reports |
| Important page not internally discoverable | No normal crawl path leads to the page | High | Crawl from the homepage and compare with sitemap/GSC URLs |
| Noncrawlable JavaScript links | Navigation acts like links for users but lacks crawlable <a href> destinations |
High | Inspect rendered HTML and link markup |
| Important page buried too deeply | Valuable page requires many navigation steps to reach | Medium–High | Measure crawl depth and inspect hierarchy |
| Crawl trap | URLs can be generated nearly infinitely | High | Review crawl patterns, parameters and log files |
| Faceted navigation crawl explosion | Filters create very large numbers of crawlable combinations | High on large sites | Crawl parameter patterns and compare indexable inventory |
| Internal search results crawlable | Search queries generate discoverable URLs with little independent value | Medium–High | Search crawl and index data for internal-search patterns |
| Calendar crawl trap | Forward/backward calendar navigation creates unlimited dated URLs | Medium–High | Inspect crawler URL patterns |
| Session IDs in URLs | Individual sessions generate alternate versions of pages | Medium | Look for recurring session parameters |
| Tracking parameters crawlable | Analytics parameters generate duplicate crawl paths | Medium | Segment URLs containing UTM or tracking parameters |
| Sort parameters crawlable | Sorting changes create additional URLs without new content value | Medium | Compare content and canonical handling |
| Filter parameters crawlable | Filter combinations multiply URL inventory | Medium–High | Crawl filters and inspect canonical/index directives |
| Important URL discovered but rarely crawled | Google knows the URL but does not request it often enough to process changes | Context–High | URL Inspection, Crawl Stats and logs |
| Sudden Googlebot crawl decline | Search engine crawling falls unexpectedly across important sections | Context–High | Compare Crawl Stats and server logs over time |
| Sudden crawl spike | Googlebot begins requesting a large number of unexpected URLs | Context | Identify patterns in Crawl Stats/logs |
| Low-value URLs dominate crawler requests | Crawl resources are spent on filters, parameters, search pages or duplicates | High on large sites | Analyze logs by URL pattern |
| XML sitemap unavailable | Search engines cannot access the submitted sitemap | Medium–High | Fetch sitemap and inspect Search Console status |
| XML sitemap malformed | XML or sitemap syntax prevents reliable processing | Medium | Validate sitemap |
| Sitemap contains redirects | Sitemap asks crawlers to process URLs that immediately move elsewhere | Medium | Crawl all sitemap URLs and segment 3xx responses |
| Sitemap contains 4xx URLs | Removed URLs remain submitted for indexing | Medium | Crawl sitemap URLs |
| Sitemap contains 5xx URLs | Submitted URLs fail at the server | High if widespread | Crawl sitemap and inspect server errors |
| Sitemap contains noindex URLs | Sitemap and indexing directives send conflicting signals | Medium | Compare sitemap URLs with robots directives |
| Sitemap contains noncanonical URLs | Submitted URLs are not the preferred versions | Medium | Compare sitemap URLs with canonical targets |
| Sitemap omits priority canonical pages | Important pages receive one fewer discovery/canonical signal | Medium | Compare indexable crawl inventory with sitemap |
Sitemap lastmod is unreliable |
Dates change without actual page changes | Low–Medium | Compare modification dates against real publishing history |
Crawling means a search engine can request a URL.
Indexing means that URL can potentially become part of the search index.
Canonicalization determines which URL Google treats as representative when several URLs contain the same or highly similar content.
Google treats canonical annotations as signals rather than absolute commands and may select a different canonical when other signals disagree.
| Issue | What it means | Typical severity | How to diagnose |
|---|---|---|---|
| Important page accidentally noindexed | A page that should appear in search explicitly requests exclusion | Critical/High | Inspect robots meta/X-Robots-Tag and URL Inspection |
| Important URL excluded unexpectedly | Page is not indexed despite being intended for search | High | Inspect Page Indexing reason and URL Inspection |
| Utility page indexed unnecessarily | Search index contains a page that was never meant to rank | Low–Medium | Compare indexed inventory against intended SEO inventory |
| Canonical points to wrong page | Signals are consolidated toward an unrelated or inferior URL | High | Crawl canonicals and manually inspect relationship |
| Canonical points to noindex | Preferred target itself requests exclusion | High | Crawl canonical destinations |
| Canonical points to redirect | Canonical requires another hop before reaching final content | Medium | Resolve canonical targets |
| Canonical points to 404/410 | Canonical target does not exist | High | Check target status |
| Canonical points to 5xx | Preferred URL is failing at server level | High | Check response and logs |
| Multiple canonical tags | Page declares more than one canonical destination | High | Inspect HTML and HTTP headers |
| HTTP-header and HTML canonicals conflict | Different implementations declare different preferred URLs | High | Compare header and document source |
| JavaScript changes canonical | Rendered canonical differs from initial HTML | Medium–High | Compare raw and rendered HTML |
| Canonical chain | Canonical A points to B, which canonicalizes to C | Medium | Resolve canonical graph |
| Malformed canonical URL | Target is invalid, incomplete, or incorrectly generated | Medium | Validate canonical syntax |
| Duplicate pages without clear canonical | Several equivalent URLs compete for representation | Medium–High | Cluster duplicate content and compare canonicals |
| Google chooses different canonical | Google's selected URL differs from declared preference | Context–High | URL Inspection |
| HTTP and HTTPS both accessible | Protocol variants duplicate site content | High if sitewide | Crawl both protocol versions |
| www and non-www both accessible | Host variants duplicate site content | High if sitewide | Request both hosts |
| Slash variants duplicate | /page and /page/ resolve independently |
Medium | Crawl both patterns |
| Case variants duplicate | /Service/ and /service/ create separate crawlable URLs |
Medium | Normalize and compare URL inventory |
| Parameter URLs indexed separately | Tracking/filter/sort variants become independent search URLs | Medium–High | Search index and crawl parameter versions |
| Print version competes with HTML | Alternate format becomes independently indexable | Medium | Check canonicals and indexed URLs |
| PDF competes with preferred HTML | PDF ranks instead of the business's intended web page | Context–Medium | SERP/GSC landing-page review |
| Staging site indexed | Development environment appears in public search | High | Search/index check and robots/authentication review |
| Old and replacement URLs both indexable | Migration or consolidation did not fully transfer | High | Compare old/new index status |
| Taxonomy index bloat | Tags, author pages or archives generate large low-value indexed inventories | Context–Medium | Crawl and compare taxonomy role |
| Soft 404 | URL returns 200 but effectively contains missing/error content | Medium–High | Search Console + manual page review |
| Canonicalized URL still heavily linked internally | Internal architecture continues telling crawlers that duplicate URL matters | Medium | Internal-link destination analysis |
A duplicate is not automatically an error. Duplicate pages can be legitimate when relationships are handled correctly, including regional variants or intentionally canonicalized versions.
Status codes tell crawlers and browsers what happened when a URL was requested.
| Issue | What it means | Typical severity | How to diagnose |
|---|---|---|---|
| Internal 404 | An internal link points to a page that does not exist | Medium–High | Crawl internal links and 4xx destinations |
| Important removed URL lacks redirect | Valuable old URL has a relevant replacement but no transfer path | High | Compare removed pages with migration/content map |
| Unintentional 410 | URL explicitly states permanent removal despite needing to remain available | High | Crawl 410 responses |
| Widespread 5xx errors | Server cannot successfully serve important pages | Critical | Crawl + server logs |
| Intermittent 5xx | Pages fail inconsistently | High | Repeated tests and logs |
| 429 blocks legitimate crawlers | Rate limiting denies search-engine requests | High | Log response codes by user agent |
| DNS failure | Domain cannot resolve correctly | Critical | DNS monitoring |
| Connection timeout | Server fails to respond within acceptable request window | High | Crawl and server monitoring |
| TLS/certificate failure | Secure connection cannot be established reliably | Critical/High | Browser/server TLS test |
| Redirect chain | URL requires multiple redirect hops | Medium | Crawl redirect paths |
| Redirect loop | Redirects cycle without reaching a final URL | High | Follow redirects |
| Broken redirect | Redirect destination returns error | High | Resolve redirect destinations |
| Temporary redirect used for permanent move | Long-term URL move is represented as temporary | Context–Medium | Review redirect intent/status |
| Permanent redirect used for temporary condition | Temporary URL state is represented as permanent | Context | Review business intent |
| Irrelevant redirect | Removed URL sends users to unrelated content | Medium–High | Manually inspect redirect relevance |
| Mass redirects to homepage | Many unrelated removed URLs funnel to homepage | Medium–High | Segment redirect destinations |
| HTTP → HTTPS redirect missing | Insecure protocol remains independently accessible | High | Request HTTP versions |
| Host redirect inconsistent | www/non-www versions do not resolve consistently | High | Test host variations |
| Redirect adds avoidable latency | Internal navigation repeatedly routes through 3xx | Medium | Crawl internal links to redirected URLs |
| Meta refresh redirect | Page-level refresh handles a move that could use server redirect | Low–Medium | Inspect HTML |
| JavaScript redirect | Critical redirect depends on script execution | Context–Medium | Test without JS |
| Sneaky mobile redirect | Mobile users receive unrelated destination | Critical | Compare mobile/desktop behavior |
| Status varies unexpectedly by user agent | Crawlers receive materially different response from normal users | High/Critical | Compare requests by user agent |
Google recommends maintaining site-move redirects for at least a year and updating internal links so they point directly to new destinations.
Google uses links both to discover pages and to understand relationships and relevance. Google recommends that every page you care about receive a link from at least one other page on your site.
| Issue | What it means | Typical severity | How to diagnose |
|---|---|---|---|
| Orphan indexable page | Page has no incoming crawlable internal links | High for important page | Compare crawl with sitemap, GSC and analytics |
| Revenue page receives few internal links | Commercial asset gets little architectural support | High | Count and assess inlinks |
| Excessive click depth | Important content sits unnecessarily far from useful hubs | Medium | Crawl-depth analysis |
| Navigation omits core page | Important service/product/location is absent from primary architecture | High | Manual navigation review |
| Hierarchy does not reflect business model | Services, products, markets or industries are organized without logical parent/child structure | High | Architecture map |
| Hub fails to link children | Parent does not expose important child entities | Medium–High | Hub-page inspection |
| Child pages lack parent path | Users/crawlers cannot easily move back through hierarchy | Medium | Breadcrumb/navigation analysis |
| Supporting content does not support commercial pages | Informational authority remains isolated | Medium–High | Contextual-link audit |
| Case studies isolated from services | Proof exists but is disconnected from claims it validates | Medium | Map case studies to services/industries |
| Mechanical sibling linking | Every page links to every sibling regardless of relevance | Low–Medium | Template-link review |
| Links point to redirects | Internal architecture uses obsolete destinations | Medium | Crawl 3xx link targets |
| Links point to 404s | Internal paths terminate at missing pages | Medium–High | Crawl link targets |
| Links point to 5xx | Internal paths lead to failing pages | High | Crawl links/status |
| Links point to noindex | Architecture promotes URLs intentionally excluded from search | Context–Medium | Compare links/directives |
| Links point to noncanonical URLs | Internal signals favor duplicate versions | Medium | Compare destinations/canonicals |
| Noncrawlable JS navigation | Clickable navigation does not expose standard crawlable links | High | Inspect rendered HTML |
| Infinite scroll lacks crawlable fallback | Deeper items cannot be reached through URLs/links | High | JS-disabled and crawler test |
| Pagination broken | Crawlers cannot traverse deeper item sets | Medium–High | Follow pagination URLs |
| Breadcrumb hierarchy wrong | Breadcrumb suggests incorrect parentage | Medium | Compare breadcrumb and architecture |
| Broken breadcrumb link | Breadcrumb leads to invalid URL | Low–Medium | Crawl breadcrumb paths |
| Breadcrumb conflicts with schema | Visible and structured breadcrumb differ | Medium | Compare HTML/schema |
| Vague anchors | Internal links use phrases such as “click here” without destination context | Low–Medium | Anchor-text report |
| Misleading anchor | Anchor promises a different topic than destination provides | Medium | Manual link review |
| Image links lack useful alt | Linked image provides little anchor context | Low–Medium | Image-link audit |
| Template/footer link overload | Large repeated link blocks reduce navigational clarity | Context | Compare link distribution |
| Important page lacks contextual links from strong pages | Valuable page receives no support from established relevant assets | Medium–High | Inlink + traffic/backlink analysis |
| URL variants created by navigation | Different components link to different versions of same destination | Medium | Normalize internal URLs |
| Internal links remain on old URLs after migration | Site continuously crawls through redirects | Medium–High | Compare inlinks against redirect map |
Google uses the <title> element, visible headings, anchor text and other signals when generating title links, and may rewrite titles that are inaccurate, boilerplate, outdated or unclear.
| Issue | What it means | Typical severity | How to diagnose |
|---|---|---|---|
| Missing title | Page lacks a basic title signal | High on indexable page | Crawl title elements |
| Duplicate titles | Distinct pages use same title | Medium | Group indexable URLs by title |
| Boilerplate titles | Titles vary only by tiny token while pages have different jobs | Medium | Template comparison |
| Inaccurate title | Title does not represent visible content | High | Compare title to page purpose |
| Keyword-stuffed title | Repetitive keywords reduce clarity and may appear spammy | Medium | Manual review |
| Outdated title | Year, product or service information is stale | Medium | Compare title to content |
Multiple <title> elements |
Browser/search engine receives competing title elements | Medium | HTML inspection |
| Title/H1 conflict | Main page signals describe different tasks | Context–Medium | Compare title, H1 and intent |
| Missing clear H1 | Main visible subject is unclear | Medium | Heading crawl |
| Several equally prominent main headings | Search/user interpretation of main title becomes ambiguous | Context–Medium | Visual inspection |
| Duplicate H1 across distinct pages | Pages may lack meaningful differentiation | Context | Compare page purpose/content |
| Poor heading hierarchy | Sections are difficult to interpret or navigate | Low–Medium | Outline page headings |
| Missing meta description | Site gives up some control over preferred snippet messaging | Low–Medium | Crawl important pages |
| Duplicate meta descriptions | Different pages use identical snippet messaging | Low | Group descriptions |
| Misleading meta description | Snippet promise differs from destination | Medium | Compare SERP snippet/page |
| Keyword-stuffed description | Snippet is written for terms rather than users | Low–Medium | Manual review |
| Poor search snippet | Generated snippet undersells or misrepresents important page | Context–Medium | SERP/GSC review |
data-nosnippet hides important text |
Useful content cannot appear in snippet | Context–Medium | Inspect attributes |
Restrictive max-snippet |
Preview controls reduce available search context | Context | Inspect robots directives |
| Missing/inaccessible favicon | Brand icon may not appear in search | Low | Check homepage favicon implementation |
| Wrong page language declaration | HTML language metadata does not match content | Low–Medium | Inspect lang attribute |
| Title uses different language/script | Title signal conflicts with primary content | Medium | Manual review |
| Site-name signals inconsistent | Homepage/Organization/site signals use different naming | Low–Medium | Compare sitewide identity |
Google's current people-first content guidance asks whether content is original, substantial, accurate, useful, clearly authored when expected, and created primarily to help an intended audience rather than attract search traffic.
| Issue | What it means | Typical severity | How to diagnose |
|---|---|---|---|
| Search-intent mismatch | Page answers a different task from what searchers expect | High | SERP analysis + query review |
| Wrong page type | Article, service, category or location page does not match query task | High | Compare ranking page types |
| Commercial demand served only by blog | Site lacks a proper buying/evaluation page | High | Query-to-page mapping |
| Informational query forced to sales page | User needs learning, not immediate conversion | Medium | SERP/intention review |
| Page fails its primary task | Visitor cannot complete the reason for arriving | High | Manual task-based review |
| Incomplete coverage | Important information required for decision/task is absent | Medium–High | Competitor + subject analysis |
| Generic content | Copy could apply to almost any competitor | Medium–High | Remove brand/name and assess uniqueness |
| Rehashed content | Page mainly summarizes existing sources without adding value | Medium–High | Source comparison |
| Factual errors | Content contains verifiably incorrect information | High | Manual verification |
| Outdated facts | Price, law, product, standard, process or statistic is obsolete | High when decision-critical | Source/date audit |
| Unsupported claims | Important performance or business statements lack evidence | Medium–High | Claim/evidence review |
| Weak YMYL trust | Sensitive advice lacks appropriate expertise/evidence | High | Expertise/source review |
| Unclear authorship | Reader cannot tell who produced expertise-heavy content | Medium | Byline/profile review |
| Weak author profile | Claimed expert page does not establish relevant background | Medium | Profile review |
| Unsubstantiated first-hand claims | Content claims experience without showing evidence | Medium | Evidence review |
| Missing examples | Technical/complex explanation never demonstrates application | Medium | Task completion review |
| Missing data/proof | Claims remain abstract despite available evidence | Medium | Evidence inventory |
| Answer buried in filler | User must read excessive setup before reaching needed answer | Medium | Editorial review |
| Repetition | Sections repeat same concept without new value | Medium | Editorial review |
| Keyword stuffing | Terms are repeated unnaturally to influence rankings | High if severe | Manual/corpus review |
| Scaled low-value content | Large content inventory was generated primarily for search manipulation | Critical/High | Template + production review |
| Doorway location pages | Similar location URLs mainly funnel visitors to same destination | High | Compare geographic pages |
| Near-duplicate service areas | City pages change names but little else | High | Similarity analysis |
| Near-duplicate industries | Vertical pages swap industry labels without industry-specific value | Medium–High | Compare templates |
| Synonym service pages | Several URLs represent same offering under different names | High | Service/entity mapping |
| Weak programmatic pages | Generated URLs offer little independent utility | High at scale | Sample template/data variation |
| Topic drift | Site publishes unrelated topics mainly to capture traffic | Medium–High | Topical inventory |
| Content decay | Once-competitive page has become stale or incomplete | Medium–High | Performance trend + content review |
| Fake freshness | Publication date changes without meaningful content improvement | Low–Medium | Revision history |
| Thin UGC indexed | Empty or low-value user pages become search inventory | Medium | UGC index audit |
| UGC spam | Users create off-topic promotional/spam pages or links | High | Search/index + moderation review |
| Supplier copy reused | Product/service descriptions duplicate source material without added value | Medium | Duplicate-content comparison |
| Commercial page lacks proof | Page makes promises without projects, testimonials, credentials or evidence | Medium–High | Conversion/trust review |
| CTA mismatches stage | Next action is too aggressive or irrelevant to visitor task | Medium | Journey review |
| Premature gating | Useful evaluation information is hidden before user can assess it | Context–Medium | UX/content review |
| Paywall implementation resembles cloaking | Search engine and legitimate users receive materially different access improperly | High | Render/access comparison |
| Wrong language for audience | Content technically targets locale but is not usable for that market | High | Locale review |
| Important subtopic absent | Page lacks a concept essential to completing its task | Medium | Semantic/content-gap analysis |
Content length alone is not an audit issue. A short page can completely satisfy a task, while a 5,000-word page can still be generic or incomplete.
Cannibalization should be diagnosed from competing tasks and search behavior, not merely because two pages mention similar keywords.
| Issue | What it means | Typical severity | How to diagnose |
|---|---|---|---|
| Two URLs target same primary task | Search engines/users have no clear preferred destination | High | Query/page map + GSC |
| Service page and article compete | Informational URL ranks for a commercial task or vice versa | High | Query/landing-page analysis |
| Synonymous service pages compete | Different names represent one underlying offering | High | Service taxonomy review |
| Location pages overlap | Geographic URLs do not represent sufficiently distinct markets/tasks | Medium–High | Page comparison + local SERPs |
| Industry page duplicates service | Vertical page adds little audience-specific information | Medium–High | Content comparison |
| Old/new page both rank | Consolidation or migration remains incomplete | High | GSC + SERP review |
| Category/tag page competes with landing page | Taxonomy page and intentional destination target same demand | Medium | Query landing pages |
| Parameters receive visibility | Duplicate URL variants appear for same queries | Medium | GSC URL filtering |
| PDF outranks HTML | Search selects document rather than intended interactive/commercial page | Context–Medium | SERP review |
| Syndicated copy outranks original | Another publisher becomes Google's preferred result | Context–High | Search exact passages |
| Multiple domains compete | Business splits authority/coverage across separate sites | Context–High | SERP/domain analysis |
| Authority split across duplicates | Backlinks and internal links point to several equivalent URLs | High | Link + canonical audit |
| Wrong URL consistently ranks | Google understands another page as better match | High | GSC page/query comparison |
| Normal ranking rotation misdiagnosed as cannibalization | Different pages appear occasionally without meaningful conflict | Context | Review sustained query/page performance |
Google can render JavaScript, but JavaScript-heavy sites introduce additional failure points. Google recommends crawlable links and warns about JavaScript-generated canonical inconsistencies and SPA soft-404 behavior.
| Issue | What it means | Typical severity | How to diagnose |
|---|---|---|---|
| Critical content exists only after JS | Initial HTML lacks main content | Context–High | Compare raw/rendered HTML |
| Critical links depend on noncrawlable JS | Users navigate but crawlers cannot reliably extract destinations | High | Rendered link inspection |
| Rendered page loses content | JS removes or fails to render important copy | High | Raw/rendered/browser comparison |
| Crawler render differs from user experience | Search engine sees materially incomplete/different page | High | URL Inspection + browser test |
| API failure prevents content | Data request fails and page remains incomplete | High | Console/network/render test |
| Essential resources blocked | JS/CSS required for page understanding cannot be fetched | High | Network/robots inspection |
| Lazy content requires interaction | Content appears only after click/scroll behavior crawlers may not execute | Medium–High | Render test |
| SPA missing route returns 200 | Nonexistent application URL behaves like valid page | High | Request fake routes |
| Client-side routing lacks crawlable URLs | Important application states have no stable URLs | High | Navigation/render audit |
| JS injects noindex | Script excludes page unintentionally | High | Rendered robots directives |
| JS changes canonical | Rendered preferred URL differs from raw HTML | High | Raw/render comparison |
| JS changes title incorrectly | Dynamic title becomes generic or wrong | Medium | Render title inspection |
| Structured data disappears after render | Client logic removes/changes markup | Medium–High | Rich Results + rendered DOM |
| Structured data conflicts with visible rendered content | Markup and final page disagree | High | Compare rendered page/schema |
| Hydration error | Framework delivers incomplete or blank page | High | Browser console + render |
| Mobile rendering loses content | Smartphone Googlebot receives less meaningful content | High | Mobile rendering comparison |
| Server and client links differ | Hydration rewrites link targets | Medium–High | Compare HTML/DOM |
| Consent blocks main content | Critical content never appears until interaction | Context–High | Clean-session render |
| Duplicate DOM content | Server/client components duplicate text/elements | Medium | Rendered DOM analysis |
| Rendering workload is unreliable | Page regularly times out/fails before important content appears | High | render timings/logs |
Current Core Web Vitals measure loading, interaction responsiveness and visual stability. Good thresholds are LCP ≤2.5 seconds, INP ≤200ms and CLS ≤0.1 at the 75th percentile of real-user page loads.
| Issue | What it means | Typical severity | How to diagnose |
|---|---|---|---|
| Poor LCP | Main content becomes visible too slowly | Medium–High | CrUX/PageSpeed field data |
| Poor INP | Interactions respond too slowly | Medium–High | CrUX + performance profiling |
| Poor CLS | Content shifts unexpectedly during use | Medium | CrUX + layout-shift inspection |
| Slow server response | Backend delays initial document delivery | Medium–High | TTFB/server monitoring |
| Render-blocking resources | CSS/JS delays visible content | Medium | Lighthouse/network waterfall |
| Excessive main-thread JS | Browser spends too long executing scripts | Medium–High | Performance profiler |
| Oversized images | Image dimensions/file sizes exceed actual need | Medium | image audit |
| Inefficient image format | File is larger than practical alternative | Low–Medium | Lighthouse/image review |
| Missing sensible lazy loading | Offscreen images load before needed | Low–Medium | network waterfall |
| LCP image lazy-loaded | Main hero image is intentionally delayed | High for affected template | inspect LCP resource |
| Fonts cause layout shift | Font swap moves visible content | Medium | CLS trace |
| Third-party scripts slow pages | Widgets/trackers consume CPU/network | Medium | waterfall + coverage |
| Weak browser caching | Repeat visits redownload stable resources | Low–Medium | response headers |
| Compression absent | Text assets transfer unnecessarily large payloads | Medium | response headers |
| Unused CSS/JS excessive | Large assets deliver code page does not use | Medium | coverage tools |
| Excessive DOM size | Very large tree increases layout/render cost | Medium | Lighthouse/DOM |
| Mobile content parity failure | Mobile version contains less important content than desktop | High | device comparison |
| Mobile structured-data mismatch | Mobile markup differs from primary desktop implementation | Medium–High | rendered source comparison |
| Mobile metadata mismatch | Smartphone version uses different titles/canonicals/directives | High | user-agent crawl |
| Mobile navigation hides important links | Important crawl paths disappear on smartphone rendering | High | mobile DOM inspection |
| Broken viewport | Page does not adapt correctly to device width | Medium | mobile test |
| Poor small-screen usability | Content/buttons/navigation are difficult to use | Medium | manual mobile QA |
| Intrusive interstitial | Overlay blocks primary content | Medium–High | mobile UX review |
| HTTPS missing | Site transmits traffic insecurely | High | crawl protocol |
| Mixed content | Secure pages load insecure assets | Medium–High | browser security report |
| Wrong mobile redirect | Smartphone lands on inappropriate URL | High | device testing |
| Ads obscure main content | Monetization interferes with page task | Medium | manual review |
| Template performance problem | Performance failure affects many valuable URLs | High | group field data by page type |
| Lab-only failure treated as user failure | Synthetic test looks poor while field experience is healthy | Context | Compare lab and CrUX |
Lab and field data can legitimately differ. Real-user field data should be evaluated separately from one synthetic test run.
| Issue | What it means | Typical severity | How to diagnose |
|---|---|---|---|
| Important image blocked | Googlebot-Image cannot request asset | Medium | robots/image request |
| Broken image | Image URL returns error | Medium | crawl image resources |
| Important image missing alt | Image lacks useful textual description/context | Medium | alt audit |
| Decorative image keyword stuffed | Alt text is used to inject keywords | Medium | manual alt review |
| Alt text duplicated mechanically | Distinct images receive identical descriptions | Low–Medium | image/alt export |
| Image requires interaction | Asset appears only after unsupported user action | Medium | render test |
| Weak image context | Page gives search engines little information about image subject | Low–Medium | surrounding copy review |
| Responsive implementation serves huge files | Mobile downloads desktop-size assets | Medium | network/dimension audit |
| Important images omitted from discovery paths | Search engines struggle to discover assets | Context | crawl/image sitemap review |
| Primary image is low quality/irrelevant | Visual asset does not support page intent | Context–Medium | editorial review |
| Video watch page noindexed | Video destination cannot appear in search | High where video matters | directives |
| Video not prominent | Video exists but page is not primarily a watch experience | Context | layout review |
| Thumbnail unavailable | Search engine cannot fetch representative image | Medium | test thumbnail |
| Video markup invalid | Required structured fields missing/wrong | Medium | Rich Results Test |
contentUrl/embedUrl inaccessible |
Google cannot reliably fetch player/content | Medium | request URLs |
| Duplicate video metadata | Multiple videos use same title/description | Medium | crawl video markup |
| Weak watch-page context | Page offers little useful supporting information | Medium | manual review |
| Stale video status | Expired/live information no longer matches reality | Medium | structured-data review |
Google can use VideoObject markup to understand attributes such as thumbnails, upload dates and duration, but eligibility depends on the page and implementation meeting video requirements.
Structured-data syntax, semantic accuracy and Google feature eligibility are three different questions.
| Issue | What it means | Typical severity | How to diagnose |
|---|---|---|---|
| Syntax error | JSON-LD/Microdata/RDFa cannot be parsed correctly | Medium | Schema validator |
| Required property missing | Page cannot qualify for intended Google feature | Medium–High | Rich Results Test |
| Useful recommended fields missing | Eligible entity is described incompletely | Low–Medium | feature documentation |
| Markup describes invisible content | Structured data claims information users cannot verify | High | compare markup/page |
| Fake entity | Markup describes business/product/job/event that does not exist | Critical/High | factual review |
| Wrong schema type | Entity class does not match real content | Medium–High | Schema.org review |
| Duplicate entity identity | Same real entity receives different @ids/data |
Medium | graph/template audit |
| Organization data inconsistent | Name, logo, URL or profiles vary by template | Medium | schema crawl |
| LocalBusiness address wrong | Structured business location conflicts with reality | High | compare GBP/site |
| Fake LocalBusiness on service-area page | Market is represented as physical branch | High | location validation |
| Product markup on category page | Collection is incorrectly represented as single Product | Medium–High | template review |
| Offer price differs from page | Markup claims different commercial terms | High | compare visible price |
| Review markup ineligible/self-serving | Markup attempts unsupported review-star use | High | policy review |
| Fake JobPosting | Markup appears where no genuine vacancy exists | Critical/High | job-page review |
| Expired job remains available | Closed role still appears current | High | dates/job inventory |
| Event dates stale | Structured event information no longer matches page | Medium–High | event audit |
| Deprecated markup maintained for vanished feature | Engineering effort targets no-longer-supported feature | Low/Context | current Google docs |
| Deprecated Schema.org type | Implementation uses outdated vocabulary where current replacement exists | Low–Medium | validator/docs |
| Invalid ProfilePage | Page does not primarily profile one entity | Medium | content/schema review |
| Certification overstated | Structured data implies qualification that does not exist | High | credential verification |
| Structured-data manual action | Google ignores markup because of policy violation | Critical | Search Console Manual Actions |
| Rich-result validity drops after deployment | Template release breaks supported markup | High | Search Console enhancements |
| Breadcrumb markup mismatch | Schema path differs from visible hierarchy | Medium | compare breadcrumb |
| Video markup mismatch | Structured video differs from actual embed | Medium–High | compare markup/video |
| Paywall markup missing/misconfigured | Search engine cannot distinguish legitimate gated content correctly | Medium | paywall tests |
| Same entity recreated inconsistently sitewide | Graph becomes ambiguous | Medium | entity-ID audit |
Google states that valid structured data does not guarantee a rich result, and policy violations can cause markup to be ignored or trigger manual actions.
| Issue | What it means | Typical severity | How to diagnose |
|---|---|---|---|
| Missing hreflang | Equivalent language/region pages do not declare relationship | Medium–High | hreflang crawl |
| Missing return link | One alternate references another without reciprocal annotation | High | cluster validation |
| Invalid language code | Annotation uses unsupported/wrong language format | Medium | validator |
| Invalid region code | Country targeting code is incorrect | Medium | validator |
| Hreflang targets noncanonical | Alternate annotation points to URL not intended for indexing | High | compare canonical |
| Hreflang targets redirect | Alternate URL moves elsewhere | Medium–High | resolve hreflang destinations |
| Hreflang targets 4xx/5xx | Alternate version is unavailable | High | crawl destinations |
| Hreflang targets noindex | Alternate cannot be indexed | High | directives |
| Incomplete language cluster | Not all equivalents are consistently represented | Medium | cluster graph |
| Missing useful x-default | Global selector/default page has no fallback where appropriate | Context | architecture review |
| Wrong locale served | Searcher reaches wrong language/market version | High | regional SERP/testing |
| Forced locale redirect | Users/crawlers cannot freely access alternate version | High | IP/browser testing |
| Only navigation translated | Main content remains wrong language | High | page comparison |
| Low-value machine translation | Localization is linguistically or commercially unusable | Medium–High | native review |
| Currency wrong | Commercial data mismatches target market | High | locale audit |
| Legal/regulatory information wrong | Page uses rules from different jurisdiction | High | expert review |
| Local contact/product information wrong | Page advertises unavailable offering/location | High | business validation |
| Canonical collapses legitimate locale variants | Language pages point to one master incorrectly | High | canonical/hreflang review |
| International architecture inconsistent | Countries/languages use mixed, confusing URL systems | Medium | URL map |
| Language switcher not crawlable | Alternate versions cannot be reached through normal links | Medium–High | link inspection |
Google says local ranking is mainly based on relevance, distance and prominence. Business Profile data should accurately represent the real business, and Google prohibits virtual offices that do not meet its physical-location requirements.
| Issue | What it means | Typical severity | How to diagnose |
|---|---|---|---|
| GBP unverified | Business has limited control/visibility over profile | High | Profile status |
| GBP suspended | Listing can no longer function normally | Critical | GBP manager |
| Duplicate profiles | Same business/location has competing listings | High | Maps search |
| Keyword-stuffed business name | Profile adds descriptors not used in real-world name | High | compare signage/site |
| Wrong primary category | Profile is classified under weaker/incorrect core business type | High | category review |
| Irrelevant additional categories | Categories describe services/features rather than actual business | Medium | category review |
| Address inaccurate | Profile location differs from actual operating location | Critical/High | documentation/site |
| Ineligible virtual office | Listing claims location business does not actually staff/operate | Critical | physical verification |
| SAB incorrectly exposes address | Service-area business displays location that should be hidden/ineligible | High | GBP configuration |
| Service area inaccurate | Listed markets do not represent actual service coverage | Medium | operational review |
| Map pin wrong | Map location points to wrong building/area | High | Maps inspection |
| Wrong phone | Searcher reaches invalid/unrelated number | High | call verification |
| Wrong website URL | Profile sends traffic to irrelevant/broken page | High | click/test URL |
| Hours inaccurate | Customers receive wrong availability information | Medium–High | business verification |
| Holiday hours stale | Profile displays wrong temporary hours | Medium | seasonal audit |
| GBP identity conflicts with site | Name/address/phone/category differ materially | High | cross-source comparison |
| Real branch lacks location page | Website does not represent genuine physical location well | Medium–High | location inventory |
| Location page NAP wrong | Page details conflict with actual branch | High | compare records |
| Branch pages not distinct | Multi-location site lacks unique information per facility | Medium | content/location review |
| Practitioner/business listings conflict | Individual professional listings duplicate or confuse organization | Context–High | Maps/entity review |
| Review spam not monitored | Fake or policy-violating feedback distorts profile | Medium | review audit |
| Incentivized review pattern | Review acquisition may violate platform policy | High | process review |
| Wrong customer Q&A | Public profile contains inaccurate answers | Medium | GBP Q&A inspection |
| Stale photos/business info | Profile no longer represents current business | Low–Medium | profile audit |
| Material citation inconsistency | Important platforms show wrong identity/location data | Medium | citation audit |
| Missing high-value citation | Relevant industry/local platform does not list business | Context–Medium | competitor/source research |
| Old address persists after move | Search ecosystem contains conflicting location data | High | citation/Maps audit |
| Closed branch remains active | Users/search engines receive outdated local entity | High | location inventory |
| Service-area page implies fake office | Website claims physical presence where none exists | High | local-page review |
| Doorway city pages | Large numbers of substantially similar city pages funnel to same offering | High | similarity/architecture audit |
| GBP landing page broken/redirecting poorly | Profile sends users through error or irrelevant destination | High | test landing URL |
Google explicitly says businesses should generally maintain one profile per business location and use accurate physical addresses/service areas.
Not every suspicious-looking backlink is a problem.
Google's disavow guidance is intentionally narrow. The tool is primarily relevant when a site has a significant pattern of artificial links and has, or is likely to receive, an unnatural-links manual action. Random spam links should not automatically become an emergency project.
| Issue | What it means | Typical severity | How to diagnose |
|---|---|---|---|
| Unnatural-link manual action | Google has manually penalized manipulative inbound linking | Critical | Search Console Manual Actions |
| Unnatural outbound-link action | Site passes ranking credit through manipulative outbound links | Critical | Search Console |
| Paid links pass credit | Links were bought for rankings without appropriate qualification | High/Critical | backlink procurement review |
| Sponsored links unqualified | Paid editorial/advertising links pass ranking credit | High | outbound/inbound contract review |
| Affiliate links unqualified | Commercial affiliate relationship is not appropriately identified | Medium–High | outbound-link audit |
| UGC links unqualified | User-added links pass normal ranking signals | Medium | forum/comment audit |
| Excessive reciprocal links | Link exchange exists primarily to manipulate rankings | High | backlink pattern analysis |
| Automated link generation | Tools create large manipulative backlink footprint | High | link acquisition history |
| Manipulative guest-post anchors | Large-scale placements use optimized commercial anchors | High | anchor/source analysis |
| Manipulative press-release anchors | Distributed releases create ranking-focused anchor links | High | backlink review |
| Suspicious sitewide external links | Templates create thousands of unrelated outbound links | Medium–High | outbound domain/link-position analysis |
| Backlinks point to 404 | Earned authority terminates on removed page | High when links valuable | broken-backlink report |
| Backlinks pass through long chains | External authority reaches destination inefficiently | Medium | resolve linked URLs |
| Valuable link lost after URL change | Migration breaks or removes referring relationship | High | historical backlink comparison |
| Linked page removed in migration | Strong external authority no longer reaches equivalent content | High | migration/backlink map |
| Authority isolated from money pages | Strong pages lack internal paths toward commercial assets | Medium–High | backlink + internal-link graph |
| Competitive page lacks authority | Page quality is strong but materially underlinked relative to SERP | Context–High | competitive link analysis |
| Link sources are irrelevant | Acquisition strategy produces unrelated placements | Medium | referring-domain review |
| Manipulative exact-match anchor concentration | Deliberate link building creates unnatural anchor pattern | High | anchor distribution + acquisition history |
| Directory links dominate authority strategy | Site depends on low-value generic listings instead of meaningful sources | Medium | referring-domain classification |
| Competitors have legitimate authority sources you lack | Relevant publications/associations/resources consistently cite competitors | Opportunity | link intersect |
| Unlinked/broken brand mention | Existing mention could link correctly but does not | Opportunity | mention monitoring |
| Outbound link points to hacked/malicious page | Your content sends users toward unsafe destination | High | outbound status/security audit |
| Broken external citation | Evidence/resource URL no longer works | Low–Medium | external-link crawl |
| Disavow file removes healthy links | Legitimate authority may be intentionally ignored | High | disavow review |
| Random spam links treated as crisis | Team wastes resources on links it did not create and Google may already ignore | Context | manual-action + acquisition-history check |
Google's spam policies specifically prohibit buying/selling links for ranking, excessive exchanges, automated link creation and paid advertorial links that pass ranking credit.
These findings can override nearly every normal SEO priority.
| Issue | What it means | Typical severity | How to diagnose |
|---|---|---|---|
| Manual action | Human Google reviewer found spam-policy violation | Critical | Search Console Manual Actions |
| Security issue | Google detects hacked/malicious behavior | Critical | Search Console Security Issues |
| Website hacked | Unauthorized party controls content/code | Critical | server/security audit |
| Code injection | Malicious scripts/code added to templates/pages | Critical | file/database diff |
| Spam page injection | Hack creates large numbers of unrelated pages | Critical | site search + crawl/logs |
| Hacked redirect | Users/bots are redirected to malicious destination | Critical | device/user-agent testing |
| Malicious canonical injection | Hacker points canonicals to external spam site | Critical | canonical crawl |
| Cloaking | Search crawlers receive materially different content to manipulate results | Critical | user-agent comparison |
| Sneaky mobile redirect | Mobile visitor is sent somewhere unrelated | Critical | mobile tests |
| Keyword stuffing | Pages repeat terms unnaturally at scale | High | content audit |
| Doorway abuse | Near-identical pages/sites exist mainly to capture query variants | High/Critical | architecture/content review |
| Scaled content abuse | Large volumes of low-value content are created primarily to manipulate rankings | High/Critical | production/template review |
| Expired-domain abuse | Expired domain's reputation is repurposed primarily for ranking manipulation | High/Critical | domain/content history |
| Site-reputation abuse | Third-party content exploits host site's ranking signals without adequate oversight | High/Critical | publishing/ownership review |
| Scraped content at scale | Site republishes others' material without meaningful added value | High | duplicate/source analysis |
| Thin affiliate inventory | Pages mainly exist to send users elsewhere without useful independent value | High | monetization/content review |
| UGC spam | Forums/profiles/comments generate off-topic spam | High | UGC crawl |
| Misleading functionality | Page promises tool/action it does not actually provide | High | functional testing |
| Malware warning | Browser/search warns users about malicious software | Critical | security reports |
| Spammy structured markup | Markup is used deceptively or violates feature policies | High | manual action/schema review |
| Pure-spam action | Site uses broad aggressive spam tactics | Critical | Manual Actions |
| Abused forum/free-host section | Third-party spam exploits site's platform | High | index/UGC review |
| Back-button hijacking | Malicious code traps users in navigation | Critical | browser behavior |
| Private/hacked URLs indexed | Sensitive or injected pages appear publicly | Critical | Search Console/site search |
Google's current spam policies cover doorway abuse, scaled content abuse, hidden text/link abuse, hacked content, site reputation abuse and other manipulative practices.
A technically healthy SEO program can still fail if the business cannot measure whether organic visibility creates qualified outcomes.
| Issue | What it means | Typical severity | How to diagnose |
|---|---|---|---|
| Analytics absent on key templates | Organic visits disappear from reporting | High | tag coverage audit |
| Analytics fires twice | Sessions/events/conversions are inflated | High | Tag Assistant/network |
| Consent implementation breaks analytics | Tracking stops for users where measurement should legally occur | Context–High | consent/tag testing |
| Wrong Search Console property | Team monitors incomplete protocol/host data | Medium | property review |
| Reporting split across host variants | Data is fragmented unnecessarily | Medium | analytics/GSC property review |
| Organic conversions untracked | Traffic cannot be tied to business actions | High | conversion inventory |
| Form submissions not measured | Lead generation is invisible | High | form-event test |
| Organic calls untracked | Local/service businesses cannot attribute phone leads | High | call tracking |
| Ecommerce revenue inaccurate | Organic sales reporting cannot be trusted | Critical/High | transaction reconciliation |
| Cross-domain journey broken | User moves between domains and session/source resets | High | cross-domain QA |
| Organic traffic misclassified | Channel rules place search visits into referral/direct/other | High | source/medium audit |
| Internal traffic pollutes data | Staff activity materially distorts smaller datasets | Medium | traffic-source review |
| Bot traffic distorts reporting | Nonhuman sessions affect performance numbers | Medium | behavior/source analysis |
| URL variants fragment reporting | Same page appears under many query-string/path versions | Medium | landing-page normalization |
| Migration has no benchmark | Team cannot determine impact of launch | High | historical data availability |
| Rankings tracked without business outcomes | Success is evaluated only by positions | High strategically | reporting review |
| CRM drops acquisition source | B2B pipeline cannot be connected to organic search | High | CRM field audit |
| Offline outcomes disconnected | Calls/closed sales never reconnect to original search source | High | attribution workflow |
| Tool health score becomes KPI | Technical warning count substitutes for business performance | Medium strategically | reporting review |
| GSC and GA4 URLs do not reconcile | URL structures/canonicals complicate comparison | Medium | normalize paths |
Site moves concentrate risk because many signals can change at once.
| Issue | What it means | Typical severity | How to diagnose |
|---|---|---|---|
| No URL migration map | Old assets lack planned destinations | Critical/High | prelaunch URL inventory |
| Old URLs not redirected | Search signals/users terminate at removed pages | Critical/High | old-URL crawl |
| Many URLs redirect generically | Specific content is mapped to homepage/category without relevance | High | redirect mapping review |
| Internal links use old URLs | New site constantly routes through redirects | Medium–High | new-site crawl |
| Canonicals reference old site | New pages consolidate to previous URLs | Critical | canonical crawl |
| Sitemap lists old URLs | Discovery signals continue promoting legacy site | High | sitemap review |
| Hreflang points to old URLs | International clusters break | High | hreflang crawl |
| Structured-data IDs reference obsolete entities/URLs | Entity graph remains tied to old architecture | Medium | schema crawl |
| Production robots blocks site | Staging crawl restriction ships live | Critical | robots test |
| Production retains noindex | Launch pages request search exclusion | Critical | directive crawl |
| Staging environment indexable | Duplicate preproduction copy becomes public | High | staging security/index check |
| Analytics lost at launch | New site traffic/conversions disappear from reporting | High | launch QA |
| Important content removed | Redesign reduces relevance/coverage | High | old/new content diff |
| Mobile content reduced | Mobile-first index receives weaker page | High | device diff |
| Navigation orphans pages | New menus remove discovery paths | High | before/after crawl |
| Metadata lost | Titles/descriptions/headings reset during template move | Medium–High | before/after comparison |
| Status codes change unexpectedly | Former live pages become errors/redirects | High | before/after URL crawl |
| Strong backlinks not reclaimed | Important external signals remain on obsolete URLs | High | backlink/redirect audit |
| Domain move lacks appropriate migration steps | Search engine needs longer to understand full domain transition | High | migration checklist |
| Redirects removed too soon | Old signals/users lose transfer path | High | legacy-URL testing |
| Framework change breaks rendering | New JS stack hides or alters content | Critical/High | raw/render comparison |
| Redesign damages performance | New layout/scripts cause real-user regressions | Medium–High | pre/post CrUX |
| Schema disappears | Template release removes valid structured data | Medium | pre/post validation |
| Hreflang disappears/breaks | Locale relationships fail after move | High | hreflang validation |
| No postlaunch anomaly monitoring | Team misses traffic/indexation failures until much later | High operationally | monitoring process |
AI visibility needs careful terminology.
There is no universal file, schema type, or shortcut that guarantees citation by AI systems.
Traditional search accessibility, clear entity information, strong source material and site authority still matter.
| Issue | What it means | Typical severity | How to diagnose |
|---|---|---|---|
| OAI-SearchBot blocked unintentionally | Site wants ChatGPT search discovery but denies the search crawler | Context–Medium | robots.txt policy review |
| Search-oriented AI crawler blocked contrary to goals | Crawler policy conflicts with desired visibility | Context | review crawler rules |
| Training and search crawlers confused | Business uses one policy for crawlers with different functions | Context | crawler/user-agent policy map |
| AI crawler rules conflict | robots rules accidentally permit/block wrong agents | Medium | robots testing |
| Critical content requires interaction | Agents/crawlers cannot retrieve core information directly | Medium | fetch/render testing |
| Entity identity inconsistent | Organization, people, products and services are described differently across pages | Medium | entity audit |
| About/identity weak | Site provides little verifiable information about who operates it | Medium | identity/trust audit |
| Important claims lack sources | Content is difficult to verify/cite reliably | Medium | evidence audit |
| Content difficult to extract | Poor structure hides primary answers, tables, definitions or relationships | Medium | document structure review |
| AI visibility measured by anecdotes | A handful of prompts are treated as statistically reliable tracking | Context | measurement methodology |
Missing llms.txt treated as defect |
Experimental/nonstandard file is presented as mandatory SEO requirement | Not an issue | challenge tool recommendation |
| AI access replaces normal indexability work | Team focuses on AI bots while core search pages remain blocked/broken | High strategically | compare priorities |
| Spam tactics used for AI search | Manipulative content/link practices are justified as “GEO” | High | content/link policy review |
| External brand/entity data inaccurate | Third-party sources describe company incorrectly | Medium | entity/brand monitoring |
Google clarified in August 2026 that its normal Search spam policies also apply to generative AI responses in Google Search.
Accessibility is not a substitute term for SEO, and an accessibility warning should not automatically be described as a ranking-factor problem.
However, several accessibility failures overlap directly with crawlability, content understanding and user experience.
| Issue | What it means | Typical severity | How to diagnose |
|---|---|---|---|
| Informative image missing alt text | Image has no textual alternative/context | Medium | accessibility/image audit |
| Linked image has empty alt | Link provides no textual anchor context | Medium | link/alt audit |
| Empty link text | Destination exists but link provides no readable context | Medium | crawler/accessibility tool |
| Buttons masquerade as links | Navigation may not expose crawlable URLs | High | HTML inspection |
| Heading hierarchy incoherent | Document structure becomes harder to interpret | Low–Medium | heading outline |
| Missing page language | Assistive technology receives weak language context | Low | HTML validation |
| Navigation inaccessible by keyboard | Some users cannot reach important content/actions | UX High / SEO indirect | keyboard test |
| Form control lacks label | Lead form becomes less usable/accessibile | Conversion issue | accessibility test |
| Duplicate element IDs | Scripts/ARIA relationships may behave unpredictably | Low | HTML validator |
| ARIA hides visible important content | Assistive tree does not represent page correctly | UX issue | accessibility inspector |
| Focus is trapped in overlay | Visitor cannot navigate past modal/interstitial | Conversion issue | keyboard test |
| Low contrast | Content is difficult for many users to read | UX issue | contrast audit |
| Autoplay/interstitial obstructs content | Primary task becomes difficult to access | Medium indirect | manual UX review |
Treat these findings accurately.
Do not tell a business:
“Color contrast is a Google ranking factor.”
Say:
“This is an accessibility and user-experience problem that should be evaluated on those grounds.”
Some of the most common audit mistakes come from fixing conditions that are functioning correctly.
A deleted URL with no relevant replacement can legitimately return 404 or 410.
The issue begins when:
valuable internal links point to it
valuable backlinks point to it
users still need it
an equivalent replacement exists
the page was removed accidentally
Redirects are necessary during:
URL changes
HTTPS migrations
domain moves
page consolidation
The issue is an unnecessary chain, loop, irrelevant destination, or redirect used where direct linking should already have been updated.
Google explicitly states that some duplicate content is normal.
The issue is often architectural:
Which URL should rank?
Are signals consolidated?
Are duplicates wasting crawling?
Are users/search engines seeing several competing destinations?
Google can generate snippets from page content.
A missing description on an important commercial page may be worth addressing.
One missing from an irrelevant utility page may not matter at all.
Modern HTML can contain multiple heading elements.
The real question is whether the page has a clear main title and understandable content hierarchy.
Google does not publish a fixed character limit for <title> elements.
Search titles are truncated as necessary based on available display space.
The problem is usually:
verbosity
repetition
weak messaging
inaccurate titles
boilerplate
not crossing an arbitrary character count.
A 250-word emergency plumbing page may answer the user's task better than a 2,500-word page filled with generic explanations.
Evaluate:
task completion
originality
evidence
coverage
clarity
business usefulness
not length alone.
nofollow, sponsored, and ugc attributes have legitimate uses.
The diagnosis depends on why the link exists and whether the attribute accurately represents the relationship.
Domain Authority, Domain Rating, Authority Score and similar metrics are third-party measurements.
They can support competitive research.
They are not Google penalties or direct Google scores.
Do not automatically upload every suspicious-looking domain into a disavow file.
First determine whether the site participated in manipulative link acquisition or has an unnatural-link problem.
Performance matters.
But improving a slow irrelevant page does not automatically deserve resources before fixing an accidentally noindexed service page.
Structured data should describe entities and supported features that actually exist.
Adding arbitrary markup for the sake of a schema score creates complexity without necessarily creating value.
llms.txt File Is Not Currently an SEO ErrorDo not classify experimental conventions as required infrastructure unless the search/AI platform you are targeting explicitly requires them.
Before creating a task, ask:
| Diagnostic question | Why it matters |
|---|---|
| Should this URL be indexed? | Many technical warnings affect URLs that should never rank |
| What job does this page perform? | Page role determines whether the issue matters |
| Does the problem affect one URL or a template? | Template failures create much greater leverage |
| Does the page already generate traffic, leads or revenue? | Existing value increases downside risk |
| Does the page have strong backlinks? | Removal/canonical errors can waste external authority |
| Does Search Console confirm an indexing/performance problem? | Crawl condition and Google behavior may differ |
| Is the issue visible to users? | Many technical problems also affect conversion or trust |
| Is this a documented requirement or a tool preference? | Prevents wasted work |
| What happens if we do nothing? | Defines actual business risk |
| Can fixing this break something already working? | Every implementation has downside risk |
The library contains both.
That distinction is important.
Something prevents an asset from functioning as intended.
Examples:
Important service page is noindexed.
Main category returns 500.
Migration redirects are missing.
Google Business Profile is suspended.
Website has a manual action.
The asset works but is underperforming.
Examples:
Important page has weak internal support.
Content no longer matches buyer intent.
Industry page contains little proof.
Product page has slow LCP.
Commercial page lacks competitive authority.
Nothing is technically broken, but improvement could create additional value.
Examples:
High-authority brand mention could become a backlink.
A valuable service page could receive contextual links from established articles.
A legitimate product filter could become a useful landing page.
A strong dataset could receive Dataset markup.
Do not report opportunities as errors.
That makes audits harder to trust.
Do not use this order:
Critical crawler warning
High crawler warning
Medium crawler warning
Low crawler warning
Use:
Fix problems threatening pages already creating valuable traffic, calls, leads, bookings, pipeline or sales.
Fix crawling, rendering, indexing, canonical, security and migration problems preventing important pages from functioning.
Resolve template-level architecture, internal-link, duplicate and inventory issues affecting many valuable pages.
Strengthen pages already showing meaningful demand and realistic opportunity.
Support priority commercial pages with stronger internal relationships, external authority, evidence and credibility.
Handle remaining warnings when the expected value justifies the effort.
The correct question is not:
How many errors can we eliminate?
It is:
Which issue is currently preventing the most valuable outcome?
Imagine a commercial contractor's crawl reports:
1,000 duplicate meta descriptions.
1 accidentally noindexed commercial service page.
The service page historically generated:
8 qualified leads per month.
The duplicate descriptions belong to:
filter and archive URLs that are not intended to rank.
A tool may visually emphasize the 1,000 warnings.
The business should fix the one noindex problem first.
That is the difference between:
auditing a website
and:
managing an SEO backlog.
A B2B engineering company deletes two pages.
No traffic
No backlinks
No internal links
Obsolete announcement from six years ago
Returning 404 is reasonable.
12 links from industry publications
Used to rank for a commercial engineering topic
Has a new equivalent service page
Returning 404 wastes an opportunity.
The same HTTP status produces two different decisions.
Context determines severity.
A home-service company has:
Boise
Meridian
Nampa
Caldwell
Each page contains the same 800 words with the city name replaced.
A crawler may report:
200 status
unique title
unique H1
self-canonical
valid meta description
Technically, every box looks green.
The architecture can still be weak.
The issue requires a human diagnosis:
Does the company genuinely serve these markets?
Does search behavior justify separate pages?
Can each page provide useful market-specific information?
Do projects, availability, pricing factors or local constraints differ?
Are these useful destinations or doorway-style variants?
Crawler health does not equal search quality.
An accounting firm publishes:
Accounting for Construction Companies
Accounting for Manufacturers
Accounting for Professional Services
All three pages use the same copy except the industry noun.
No crawler error exists.
The audit issue is:
Weak industry differentiation.
Diagnosis requires examining:
buyer problems
terminology
workflows
regulatory context
financial requirements
proof
case studies
services used
decision criteria
The problem is semantic and commercial, not technical.
A backlink tool reports:
3,800 potentially toxic links.
Before taking action, check:
Did the business create or buy those links?
Did a prior agency run link schemes?
Does Search Console show an unnatural-links manual action?
Are the links simply scraper/spam domains created without the company's involvement?
Google warns that disavowing legitimate backlinks can harm performance and says most sites do not need to use the disavow tool routinely.
A backlink toxicity score is evidence to investigate.
It is not a penalty notice.
Once issues are validated, put them into one decision system.
| Field | Purpose |
|---|---|
| Issue | What was found |
| URL / asset | What it affects |
| Issue category | Technical, content, links, local, measurement, etc. |
| Page type | Service, industry, location, product, guide, profile, etc. |
| Business role | Why the asset exists |
| Search evidence | GSC, SERP, crawl, backlinks, logs |
| Current performance | Existing traffic/rankings/conversions |
| Severity | Potential technical consequence |
| Business impact | Effect on revenue/customer outcomes |
| Scale | One URL, template, section or whole site |
| Confidence | Strength of diagnosis |
| Recommended action | What should change |
| Owner | SEO, content, dev, design, local, analytics |
| Dependency | What must happen first |
| Status | Validate, Now, Next, Backlog, Rejected |
| Result | What changed after implementation |
This forces a 404 to compete against a content gap.
A title issue competes against a broken migration.
A backlink opportunity competes against a service-page improvement.
That is how audit findings become strategy.
The most serious issues are generally those that prevent important pages or entire websites from being crawled, rendered, indexed or correctly consolidated, or that expose the site to security or spam penalties.
Examples include accidental sitewide noindex directives, major server failures, broken migrations, incorrect canonicals across important templates, hacked websites and manual actions.
Severity still depends on which assets are affected.
No.
Many audit findings are warnings, intentional configurations or conditions that do not affect valuable pages.
Validate the intended behavior before creating work.
An issue means something is preventing an asset from functioning as intended.
An opportunity means the asset works but could create more value with additional improvement.
For example:
A noindexed service page is an issue.
A service page ranking position eight that could benefit from stronger internal links is an opportunity.
Prioritize technical issues by what they prevent.
An issue blocking a commercial page from indexing should usually come before a minor metadata warning on hundreds of low-value URLs.
Evaluate business value, search evidence, scale, severity, confidence and effort together.
Duplicate content by itself is not automatically a spam violation.
Duplicate URLs become problematic when they create unclear canonicalization, split signals, waste crawling, confuse users, or represent manipulative doorway/scaled-content strategies.
Not automatically.
A 404 is appropriate when a page is permanently gone and has no relevant replacement.
It becomes important when valuable internal links, backlinks, traffic or a legitimate replacement are involved.
Redirect chains add unnecessary requests and complicate crawling and user navigation.
They should usually be simplified where practical, especially when internal links can point directly to the final destination.
Usually not by themselves.
Google can generate search snippets from page content.
Meta descriptions deserve more attention on valuable pages where better snippet messaging can improve how the result communicates its relevance.
Not necessarily.
The real issue is whether the page has a clear primary heading and understandable information structure.
Do not create high-priority engineering tickets based solely on an arbitrary “one H1 only” rule.
No.
Thinness should be evaluated based on usefulness, originality, task completion, evidence and differentiation.
There is no universal minimum word count that makes a page valuable.
No.
Structured data should accurately represent entities and content that exist and, where relevant, support search features that apply to the page.
More schema types do not automatically create better SEO.
Treat third-party toxicity scores as investigation signals, not Google penalties.
Review acquisition history, link patterns and Search Console before removing or disavowing backlinks.
Monitoring should be continuous for major anomalies.
A broader audit cadence depends on site size and change frequency.
Websites with frequent releases, large ecommerce inventories, many locations, active publishing or ongoing migrations need more frequent auditing than a small stable business website.
No single tool is enough.
A strong audit can combine:
a crawler
Google Search Console
analytics
PageSpeed/CrUX
server logs where appropriate
backlink data
Google Business Profile for local businesses
structured-data validators
manual search-result review
content analysis
and business performance data.
Each tool uses its own rules and scoring system.
A crawler can estimate the technical seriousness of a condition.
It cannot know the commercial value of the affected page, whether the behavior is intentional, or what the business should prioritize.
That requires diagnosis.
A crawler can tell you:
A canonical exists.
It cannot always tell you whether the canonical points to the right business entity.
A crawler can tell you:
Two pages have similar text.
It cannot decide whether those pages serve legitimately different buyers.
A backlink tool can tell you:
A domain looks suspicious.
It cannot know whether your company intentionally participated in a link scheme.
A speed test can tell you:
A template is slow.
It cannot tell you whether fixing that template is more valuable than restoring the service page producing your leads.
An audit becomes useful when technical findings are connected to:
Search demand
Page purpose
Business value
Customer intent
Authority
Conversions
Revenue
That is why this library separates detection from diagnosis and severity from priority.
Find the issue.
Confirm what it actually affects.
Understand why it is happening.
Then decide whether fixing it deserves resources before everything else.
Diakachimba audits technical performance, content, architecture, search coverage, local visibility, authority, links and measurement as one system.
The goal is not to hand you a spreadsheet with hundreds of warnings.
It is to identify which issues are suppressing the outcomes your business cares about and determine what should be fixed first.