Seo Forensic Incident Response

Incident Description

• When did you first notice the drop?
• Was it sudden (1–3 days) or gradual (weeks)?
• Which metrics are affected? (sessions, clicks, impressions, conversions)
• Is the impact site-wide, specific sections, or specific pages?

Incident Classification Framework

Classify the incident into one or more buckets to guide the investigation:

1. Algorithm / Core Update Impact

   • Drop coincides with known Google core update dates
   • Impact skewed toward certain types of queries or content
   • No major technical changes around the same time

2. Technical / Infrastructure Failure

   • Indexing/crawlability suddenly impaired
   • Widespread 5xx/4xx errors
   • Robots.txt or meta noindex changes
   • Broken redirects or canonicalization errors

3. Manual Action / Policy Violation

   • Manual action message in GSC
   • Sudden, severe drop in branded and non-branded queries
   • History of aggressive link building or spammy tactics

4. Content / Quality Reassessment

   • Specific sections or topics hit harder
   • Content thin, outdated, or heavily AI-generated
   • Competitors significantly improved content around the same topics

5. Demand / Seasonality / External Factors

   • Search demand drop in the niche (check industry trends)
   • Macro events, regulation changes, or market shifts

Data-Driven Investigation Steps

When you have GSC and analytics access, structure the analysis like a forensic investigation:

1. Timeline Reconstruction

• Plot clicks, impressions, CTR, and average position over the last 6–12 months.
• Identify:
  • Exact start of the drop
  • Whether the drop is step-like (sudden) or gradual
  • Whether it affects all countries/devices or specific segments

Use this to narrow likely causes:

• Step-like drop → technical issue, manual action, deployment.
• Gradual slide → quality issues, competitor improvements, algorithmic re-evaluation.

2. Segment Analysis

Segment the impact by:

• Device: desktop vs. mobile
• Country / region
• Query type: branded vs. non-branded
• Page type: home, category, product, blog, docs, etc.

Look for patterns:

• Only mobile affected → potential mobile UX, CWV, or mobile-only indexing issue.
• Specific country affected → geo-targeting, hreflang, local factors.
• Non-branded hit harder than branded → often algorithm/quality-related.

3. Page-Level Impact

Identify:

• Top pages with largest drop in clicks and impressions.
• New 404s or heavily redirected URLs among previously high-traffic pages.
• Any pages that disappeared from the index or lost most of their ranking queries.

Check for:

• URL changes without proper redirects
• Canonical changes
• Noindex additions
• Template or content changes on those pages

4. Technical Integrity Checks

Focus on incident-related technical regressions:

• Robots.txt

  • Any recent changes?
  • Are key sections blocked unintentionally?

• Indexation & Noindex

  • Sudden spike in “Excluded” or “Noindexed” pages in GSC
  • Important pages with meta noindex or X-Robots-Tag set incorrectly

• Redirects

  • New redirect chains or loops
  • HTTP → HTTPS consistency
  • www vs. non-www consistency
  • Migrations without full redirect mapping

• Server & Availability

  • Increased 5xx/4xx in logs or GSC
  • Downtime or throttling by security tools
  • Rate-limiting or blocking of Googlebot

• Core Web Vitals (CWV)

  • Sudden degradation in CWV affecting large portions of the site
  • Especially on mobile

5. Content & Quality Reassessment

When technical is clean, analyze content factors:

• Which topics or content types were hit hardest?
• Is content:
  • Thin, generic, or outdated?
  • Over-optimized or keyword-stuffed?
  • Lacking original data, examples, or experience?

Evaluate against E-E-A-T:

• Experience: Does the content show first-hand experience?
• Expertise: Is the author qualified and clearly identified?
• Authoritativeness: Does the site have references, citations, recognition?
• Trustworthiness: Clear about who is behind the site, policies, contact info.

Forensic Hypothesis Building

Use a hypothesis-driven approach instead of listing random issues.

For each plausible cause:

• Hypothesis: e.g., “A recent deployment introduced noindex tags on key templates.”
• Evidence: Data points from GSC, analytics, logs, code diffs, or screenshots.
• Impact: Which sections/pages are affected and by how much.
• Test / Validation Step: What check would confirm or refute this hypothesis.
• Suggested Fix: Concrete remediation action.

Prioritize hypotheses by:

1. Severity of impact
2. Ease of validation
3. Reversibility (how easy it is to roll back or adjust)

Output Format

Structure your final forensic report clearly:

Executive Incident Summary

• Incident type classification (technical, algorithmic, manual action, mixed)
• Date range of impact and severity (approximate % drop)
• Top 3–5 likely root causes
• Overall confidence level (Low/Medium/High)

Evidence-Based Findings

For each key finding, include:

• Finding: Short description of what is wrong.
• Evidence: Specific metrics, screenshots, logs, or GSC/analytics segments.
• Likely Cause: How this could lead to the observed impact.
• Impact: High/Medium/Low.
• Fix: Concrete, implementable recommendation.

Prioritized Action Plan

Break down into phases:

1. Critical Immediate Fixes (0–3 days)

   • Issues that block crawling, indexing, or basic site availability.
   • Reversals of harmful recent deployments.

2. Stabilization (3–14 days)

   • Clean up redirects, canonicals, internal links.
   • Restore or improve critical content and templates.

3. Recovery & Hardening (2–8 weeks)

   • Content quality improvements.
   • E-E-A-T enhancements.
   • Technical hardening to prevent recurrence.

4. Monitoring Plan

   • Metrics and dashboards to watch.
   • Checkpoints to assess partial recovery.
   • Criteria for closing the incident.

Task-Specific Questions

When helping a user, ask:

1. When exactly did you notice the drop? Any change logs around that date?
2. Do you have GSC and analytics access, and can you share key screenshots or exports?
3. Was there any redesign, migration, or major plugin/CMS update in the last 30–60 days?
4. Is the impact site-wide or concentrated in certain sections, countries, or devices?
5. Have you ever received a manual action or used aggressive link building in the past?

Related Skills

• seo-audit: For general SEO health checks outside of incident scenarios.
• ai-seo: For optimizing content for AI search experiences.
• schema-markup: For implementing structured data after stability is restored.
• analytics-tracking: For ensuring measurement is correct post-incident.

Limitations

• Use this skill only when the task clearly matches the scope described above.
• Do not treat the output as a substitute for environment-specific validation, testing, or expert review.
• Stop and ask for clarification if required inputs, permissions, safety boundaries, or success criteria are missing.