{"slug":"panda-quality-score-audit","title":"Panda Quality Score Audit — Patent-Grounded Content Quality Assessment","tags":["seo","content-quality","panda","google-updates","eeat"],"agent_summary":"Audits pages against Google's Panda patent family (US9135307B2, US9031929B1) covering content uniqueness, ad-to-content ratio, expert authorship, referral ratios, bounce pad detection, site-level quality propagation, and user engagement.","trigger_phrases":["Panda audit","quality score audit","Google Panda","thin content audit","content quality patent","site quality score","Panda risk","quality classifier"],"runnable":true,"markdown":"\n## Source and Patent Foundation\n\nGoogle's Panda algorithm (launched 2011) is a pre-query quality classifier that scores content before any search query is matched. Two patents define the signals:\n\n- **US9135307B2** — \"Ranking search results\" (Panda Algorithm)\n- **US9031929B1** — \"Site quality score\" (Navneet Panda, site-level quality)\n\nSource analysis: Bill Slawski, SEO by the Sea. The classifier is trained on human quality rater judgments and uses a decision tree to extract measurable features from pages.\n\n## How Panda Thinks\n\n**Panda is pre-query.** Quality scores are computed before any search query is matched. The page either passes or it does not, regardless of what keyword it targets.\n\n**Panda operates at both page AND site level.** A single bad page does not kill a site, but a high ratio of bad pages does. Good pages on a bad site get dragged down.\n\n**Panda uses ratios, not absolute numbers.** A 50-page site with 45 quality pages is healthier than a 5,000-page site with 500 quality pages, even though the larger site has more good content.\n\n**The decision tree is sequential.** Content uniqueness is evaluated first. If a page fails uniqueness, downstream signals matter less.\n\n## The Panda Decision Tree (Audit Checklist)\n\n### Stage 1: Uniqueness (First Gate)\n\nDoes this page contain content that could not be trivially replicated by a template, scraper, or content spinner?\n\n- Is at least 70% of the content original? (Not boilerplate, not scraped, not stock descriptions)\n- Does the content contain specific data points, case studies, or original analysis not found elsewhere?\n- Is the page clearly differentiated from similar pages on the same site?\n\n**Fail here = page is classified low quality regardless of downstream signals.**\n\n### Stage 2: Content Depth\n\n- Does the page go deeper than a basic overview?\n- Are there numeric values, specific examples, citations, or expert opinions?\n- Does the page satisfy the full scope implied by its title?\n\n### Stage 3: Ad-to-Content Ratio (Above-the-Fold Patent WO2014209758A1)\n\n- Is the primary content visible above the fold without scrolling?\n- Do ads occupy less than 40% of above-the-fold space?\n- Is the content-to-navigation ratio high on mobile?\n\n### Stage 4: Expert Authorship (Author Reputation Patent US8150842B2)\n\n- Is the author identified by name?\n- Does the author have verifiable credentials in the topic area?\n- Are there citations from authoritative institutions?\n- Is the author associated with multiple quality pages on the same topic?\n\n### Stage 5: Referral-to-Page Ratio\n\n- Does the page receive editorial links (not paid, not directories)?\n- Is the link-to-content ratio proportional? (A page with 5 inbound links and 5,000-word depth scores differently than a page with 500 inbound links but 200 words)\n\n### Stage 6: User Engagement (Dwell Time Patent US9558233B1)\n\n- Does the page keep users engaged (low bounce rate for the query type)?\n- Do users return to the SERP quickly after visiting (pogo-sticking)?\n- What is the average session depth from this page?\n\n### Stage 7: Bounce Pad Detection (US8707459B2)\n\n- Does the page exist primarily to rank and redirect to another page?\n- Is the content a thin gateway to affiliate links or ads?\n- Does clicking through the page add any original value?\n\n### Stage 8: Site-Level Quality Propagation\n\n- What percentage of the site's pages would pass Stages 1-7?\n- Are there entire sections (e.g., auto-generated location pages) that fail uniqueness?\n- Do those sections suppress the ranking ability of quality pages in other sections?\n\n## Scoring and Remediation\n\n| Stage | Score | Remediation |\n|-------|-------|-------------|\n| Uniqueness | Pass/Fail | Rewrite or remove duplicate content |\n| Content Depth | 1-5 | Add data, examples, expert citations |\n| Ad Ratio | Pass/Fail | Reduce above-fold ads, push content up |\n| Expert Authorship | 1-5 | Add author bio, credentials, citations |\n| Referral Ratio | 1-5 | Build topical links, remove thin link magnets |\n| User Engagement | 1-5 | Improve match between content and query intent |\n| Bounce Pad | Pass/Fail | Expand thin gateway pages or remove |\n| Site Level | % quality | Remove or noindex failing sections |\n\nSite-level cleanup rule: if more than 30% of crawlable pages would fail Stage 1 (uniqueness), begin a consolidation and noindex campaign before any other optimization.\n\n## Cross-References\n\n- `eeat-audit` — E-E-A-T scoring for author authoritativeness dimension\n- `google-update-proofing` — broader update-proofing strategy including KGR and Avalanche Theory\n- `helpful-content-audit` — Google's own content self-assessment checklist\n- `duplicate-content-risk-scanner` — identifying uniqueness failures at scale\n\n#seo-sop #seo #content-quality #panda #google-updates #eeat\n","html":"<h2>Source and Patent Foundation</h2>\n<p>Google's Panda algorithm (launched 2011) is a pre-query quality classifier that scores content before any search query is matched. Two patents define the signals:</p>\n<ul>\n<li><strong>US9135307B2</strong> — \"Ranking search results\" (Panda Algorithm)</li>\n<li><strong>US9031929B1</strong> — \"Site quality score\" (Navneet Panda, site-level quality)</li>\n</ul>\n<p>Source analysis: Bill Slawski, SEO by the Sea. The classifier is trained on human quality rater judgments and uses a decision tree to extract measurable features from pages.</p>\n<h2>How Panda Thinks</h2>\n<p><strong>Panda is pre-query.</strong> Quality scores are computed before any search query is matched. The page either passes or it does not, regardless of what keyword it targets.</p>\n<p><strong>Panda operates at both page AND site level.</strong> A single bad page does not kill a site, but a high ratio of bad pages does. Good pages on a bad site get dragged down.</p>\n<p><strong>Panda uses ratios, not absolute numbers.</strong> A 50-page site with 45 quality pages is healthier than a 5,000-page site with 500 quality pages, even though the larger site has more good content.</p>\n<p><strong>The decision tree is sequential.</strong> Content uniqueness is evaluated first. If a page fails uniqueness, downstream signals matter less.</p>\n<h2>The Panda Decision Tree (Audit Checklist)</h2>\n<h3>Stage 1: Uniqueness (First Gate)</h3>\n<p>Does this page contain content that could not be trivially replicated by a template, scraper, or content spinner?</p>\n<ul>\n<li>Is at least 70% of the content original? (Not boilerplate, not scraped, not stock descriptions)</li>\n<li>Does the content contain specific data points, case studies, or original analysis not found elsewhere?</li>\n<li>Is the page clearly differentiated from similar pages on the same site?</li>\n</ul>\n<p><strong>Fail here = page is classified low quality regardless of downstream signals.</strong></p>\n<h3>Stage 2: Content Depth</h3>\n<ul>\n<li>Does the page go deeper than a basic overview?</li>\n<li>Are there numeric values, specific examples, citations, or expert opinions?</li>\n<li>Does the page satisfy the full scope implied by its title?</li>\n</ul>\n<h3>Stage 3: Ad-to-Content Ratio (Above-the-Fold Patent WO2014209758A1)</h3>\n<ul>\n<li>Is the primary content visible above the fold without scrolling?</li>\n<li>Do ads occupy less than 40% of above-the-fold space?</li>\n<li>Is the content-to-navigation ratio high on mobile?</li>\n</ul>\n<h3>Stage 4: Expert Authorship (Author Reputation Patent US8150842B2)</h3>\n<ul>\n<li>Is the author identified by name?</li>\n<li>Does the author have verifiable credentials in the topic area?</li>\n<li>Are there citations from authoritative institutions?</li>\n<li>Is the author associated with multiple quality pages on the same topic?</li>\n</ul>\n<h3>Stage 5: Referral-to-Page Ratio</h3>\n<ul>\n<li>Does the page receive editorial links (not paid, not directories)?</li>\n<li>Is the link-to-content ratio proportional? (A page with 5 inbound links and 5,000-word depth scores differently than a page with 500 inbound links but 200 words)</li>\n</ul>\n<h3>Stage 6: User Engagement (Dwell Time Patent US9558233B1)</h3>\n<ul>\n<li>Does the page keep users engaged (low bounce rate for the query type)?</li>\n<li>Do users return to the SERP quickly after visiting (pogo-sticking)?</li>\n<li>What is the average session depth from this page?</li>\n</ul>\n<h3>Stage 7: Bounce Pad Detection (US8707459B2)</h3>\n<ul>\n<li>Does the page exist primarily to rank and redirect to another page?</li>\n<li>Is the content a thin gateway to affiliate links or ads?</li>\n<li>Does clicking through the page add any original value?</li>\n</ul>\n<h3>Stage 8: Site-Level Quality Propagation</h3>\n<ul>\n<li>What percentage of the site's pages would pass Stages 1-7?</li>\n<li>Are there entire sections (e.g., auto-generated location pages) that fail uniqueness?</li>\n<li>Do those sections suppress the ranking ability of quality pages in other sections?</li>\n</ul>\n<h2>Scoring and Remediation</h2>\n<p>| Stage | Score | Remediation |\n|-------|-------|-------------|\n| Uniqueness | Pass/Fail | Rewrite or remove duplicate content |\n| Content Depth | 1-5 | Add data, examples, expert citations |\n| Ad Ratio | Pass/Fail | Reduce above-fold ads, push content up |\n| Expert Authorship | 1-5 | Add author bio, credentials, citations |\n| Referral Ratio | 1-5 | Build topical links, remove thin link magnets |\n| User Engagement | 1-5 | Improve match between content and query intent |\n| Bounce Pad | Pass/Fail | Expand thin gateway pages or remove |\n| Site Level | % quality | Remove or noindex failing sections |</p>\n<p>Site-level cleanup rule: if more than 30% of crawlable pages would fail Stage 1 (uniqueness), begin a consolidation and noindex campaign before any other optimization.</p>\n<h2>Cross-References</h2>\n<ul>\n<li><code>eeat-audit</code> — E-E-A-T scoring for author authoritativeness dimension</li>\n<li><code>google-update-proofing</code> — broader update-proofing strategy including KGR and Avalanche Theory</li>\n<li><code>helpful-content-audit</code> — Google's own content self-assessment checklist</li>\n<li><code>duplicate-content-risk-scanner</code> — identifying uniqueness failures at scale</li>\n</ul>\n<p>#seo-sop #seo #content-quality #panda #google-updates #eeat</p>\n"}