| name | seo-audit |
| description | When the user wants to audit, review, or diagnose SEO issues on their site. Also use when the user mentions 'SEO audit,' 'technical SEO,' 'why am I not ranking,' 'SEO issues,' 'on-page SEO,' or 'site audit.' |
| allowed-tools | Read, Write, WebSearch, WebFetch, AskUserQuestion |
| model | sonnet |
SEO Audit
Expert knowledge for diagnosing technical SEO issues, on-page problems, and content gaps that prevent sites from ranking.
Audit Framework
An SEO audit covers five layers. Work through them in order — technical issues block all subsequent gains.
1. Crawlability → 2. Indexation → 3. On-Page → 4. Content → 5. Authority
Don't optimize content on pages that aren't being crawled correctly.
Layer 1: Crawlability
robots.txt
Check /robots.txt for:
- Blocked URLs that should be indexed (especially
/blog/, /pricing/, /product/)
- Blocked directories that contain important pages
- Disallowed user-agents that should be allowed
Critical: Disallow: / blocks the entire site. This is a catastrophic error that happens after migrations.
Tool: Fetch yoursite.com/robots.txt directly in browser.
XML Sitemap
Requirements:
- Sitemap exists at
/sitemap.xml or /sitemap_index.xml
- All important URLs included
- No noindex URLs in the sitemap (contradictory signals)
lastmod dates present and accurate- Sitemap submitted to Google Search Console
Check: Does GSC show sitemap submitted and processed?
Crawl Errors
In Google Search Console → Coverage report:
- Excluded by robots.txt: Are important pages excluded?
- Blocked by noindex: Intentional or accidental noindex tags?
- Crawl anomalies: Server errors during crawl
- Not found (404): Pages generating 404 errors (check if they have backlinks)
Tool: Screaming Frog SEO Spider for full crawl (free up to 500 URLs).
Page Speed and Core Web Vitals
Google uses CWV as a ranking signal. Measure:
| Metric | Good | Needs Work | Poor |
|---|
| LCP (Largest Contentful Paint) | < 2.5s | 2.5-4s | > 4s |
| FID / INP (Interaction to Next Paint) | < 200ms | 200-500ms | > 500ms |
| CLS (Cumulative Layout Shift) | < 0.1 | 0.1-0.25 | > 0.25 |
Tools: Google PageSpeed Insights, Core Web Vitals report in GSC, web-vitals Chrome extension.
Common CWV problems and fixes:
- LCP: Large unoptimized images → compress and convert to WebP; add
loading="lazy" to below-fold images; add fetchpriority="high" to hero image
- CLS: Images without width/height attributes → add dimensions; fonts causing layout shift → use
font-display: swap
- INP: Heavy JavaScript blocking interaction → defer non-critical JS
Layer 2: Indexation
Google Search Console Coverage
Categories to review:
Valid (indexed): Are the right pages indexed?
- Do you have pages indexed that shouldn't be? (admin pages, thank-you pages, filtered URLs)
- Are important pages missing from indexed?
Excluded:
- "Crawled, currently not indexed" → Google crawled but chose not to index. Usually: thin content, duplicate content, or low quality signal.
- "Discovered, currently not indexed" → Google knows about the page but hasn't crawled it yet. Usually: low PageRank signal, crawl budget issues.
- "Duplicate, submitted URL not selected as canonical" → You submitted a URL but Google chose a different canonical.
Canonical Tags
Check for:
- Canonical tag present on every page (self-canonical)
- Canonical points to the correct, preferred URL
- No pages pointing canonical to a different page (unintended)
- Consistent canonical form (www vs non-www, trailing slash vs no trailing slash)
Common mistake: Paginated pages all pointing to page 1 as canonical. This tells Google to ignore pages 2-N.
Duplicate Content
URL parameter duplication:
- Does
/page?color=red and /page?color=blue produce identical content?
- If yes, canonicalize or noindex the parameterized versions
HTTP/HTTPS duplication:
- Does your site serve on both http:// and https://?
- All http pages should 301 redirect to https
www/non-www duplication:
- Does your site serve on both www and non-www?
- Pick one; redirect the other
Check with: Screaming Frog crawl + duplicate content analysis; Copyscape for external duplicate detection.
Layer 3: On-Page SEO
Title Tags
Requirements:
- Unique on every page
- Contains the primary keyword (near the front preferred)
- 50-60 characters (displays fully in SERPs)
- Each title should be descriptively different from others
Common issues:
- Duplicate titles ("Home | Company Name" repeated across pages)
- Missing titles (defaults to page URL or H1)
- Too long (truncated in SERPs at ~60 chars)
- Not keyword-optimized (just brand name or generic terms)
Meta Descriptions
Requirements:
- Unique on every page
- 150-160 characters
- Contains a call to action and the primary keyword
- Accurately describes the page (Google may rewrite if not accurate)
Note: Meta descriptions are not a direct ranking factor, but affect CTR.
H1 Tags
Requirements:
- One H1 per page (not zero, not multiple)
- Contains the primary keyword
- Describes the page's core topic
- Different from the title tag (not identical)
Heading Hierarchy
- H1 → H2 → H3 structure (don't skip levels)
- H2s should cover major sections
- H3s break down subsections
- Use keywords naturally in H2s and H3s (subkeywords and related terms)
Internal Linking
Audit questions:
- Do all important pages have internal links pointing to them?
- Are any important pages "orphaned" (no internal links)?
- Are anchor texts descriptive (not "click here" or "read more")?
- Are there broken internal links (Screaming Frog → Response Codes → 4xx)?
PageRank distribution: Internal links pass PageRank. Important pages need more internal links from high-authority pages.
Layer 4: Content Audit
E-E-A-T Signals
E-E-A-T = Experience, Expertise, Authoritativeness, Trustworthiness. Not a direct algorithm, but a quality evaluation framework Google uses for manual reviews and training its systems.
Experience signals:
- Author bylines with relevant experience stated
- First-hand accounts and case examples
- Original data and research
Expertise signals:
- Author credentials relevant to the topic
- Depth and accuracy of content
- Citations of reputable sources
Authoritativeness signals:
- Backlinks from authoritative sites in your space
- Brand mentions across the web
- Press coverage
Trustworthiness signals:
- HTTPS
- Clear privacy policy and contact information
- Real company address and phone number
- No broken links or pages
- Accurate, up-to-date content
Content Quality Assessment
For each key page, evaluate:
| Criterion | Check |
|---|
| Search intent match | Does this content match what someone searching this keyword actually wants? |
| Comprehensiveness | Does it cover the topic thoroughly? |
| Freshness | Is the content current? Are examples and stats recent? |
| Unique value | Does this add something not found elsewhere? |
| Engagement signals | Low bounce rate, longer time-on-page |
"Crawled, currently not indexed" pages often have thin content. They need expansion before indexing.
Keyword Cannibalization
Occurs when multiple pages target the same keyword. Google gets confused about which to rank.
Detection: Use Ahrefs or Semrush site explorer → search for your target keyword → see if multiple pages rank for it.
Fix: Consolidate into one comprehensive page, or clearly differentiate intent between pages.
Layer 5: Authority (Backlink Profile)
Backlink Audit
Tools: Ahrefs Site Explorer, Semrush Backlink Analytics, Google Search Console (Links report)
Key metrics:
- Domain Rating (Ahrefs DR) or Domain Authority (Moz DA)
- Total referring domains (more unique domains = better)
- Anchor text distribution (over-optimized exact-match anchors are a red flag)
- Lost backlinks (identify if major links were recently lost)
Toxic link signals:
- Links from link farms or PBNs (Private Blog Networks)
- Links from sites with DR < 10 and no organic traffic
- Sitewide footer links with exact-match anchor text
- Sudden spike in links from unrelated foreign-language sites
Disavow: Use Google's disavow tool sparingly and only for clearly toxic links you cannot get removed. Wrong disavow submissions can harm your site.
Audit Report Structure
Organize findings into:
Critical (fix within 1 week): Issues that are actively harming rankings or blocking indexation
- Crawl blocks on important pages
- Whole-site noindex or robots block
- Canonical pointing wrong
- 5xx server errors
High (fix within 1 month): Issues with significant ranking impact
- CWV fails
- Duplicate title tags
- Missing or duplicate H1s
- Key pages not indexed
Medium (fix within quarter): Issues that add up over time
- Missing meta descriptions
- Thin content on secondary pages
- Missing alt text on images
- Broken internal links
Low (optimize when possible):
- Suboptimal title tag lengths
- Missing schema markup
- Image file size optimization
Common Rationalizations
| Rationalization | Reality |
|---|
| "We rank for our brand — SEO is fine" | Brand rankings are easy. Non-branded organic traffic is the revenue driver. |
| "We did an SEO audit 2 years ago" | SEO changes. Google algorithm updates, site changes, and competitive shifts require annual audits at minimum. |
| "Our developer handles technical SEO" | Developers fix technical issues; they don't diagnose SEO strategy. Both roles are needed. |
| "We can't fix the Core Web Vitals — it requires a site rebuild" | Most CWV improvements are targeted changes (image optimization, JS deferral) not full rebuilds. |
| "We don't have time to write new content" | Before writing new content, audit and improve existing content. Consolidating thin pages often outperforms new pages. |
Verification
