Content Freshness & Decay Monitor
Patents: US 7,346,839 — "Information retrieval based on historical data" (Matt Cutts, Paul Haahr) + US 8,549,014 B2 — Content Freshness Scoring Source: Bill Slawski, SEO by the Sea
Core Concept
Google builds a temporal model for every indexed page. Based on the content type, topic category, and historical update patterns, Google assigns an "expected freshness" — how recently content should have been updated to be considered authoritative. When content falls behind its expected freshness, its ranking signal degrades over time.
This is not about whether content is "recent." It's about whether the content is as fresh as its topic requires.
The 4 Content Classes
Classify every page before auditing decay risk.
Class 1: Evergreen
Description: Static technical facts, foundational definitions, reference guides, how-to fundamentals on stable topics Expected freshness: Years. Content can remain unchanged without decay penalty. Decay risk: Very low. A "what is TCP/IP" article from 2018 is still accurate and ranks fine. Trigger for update: When underlying topic actually changes (not calendar-based)
Examples: Grammar rules, mathematical formulas, historical events, technical API documentation for stable APIs, foundational SEO concepts
Class 2: Semi-Evergreen
Description: Best practices, tool guides, strategy frameworks, industry-standard methodologies Expected freshness: 6-18 months Decay risk: Moderate. Tool UIs change, best practices evolve, industry standards shift. Trigger for update: When tools update, best practices shift, or competitor content is fresher
Examples: SEO audit checklists, social media strategy guides, marketing tool tutorials, CMS setup guides
Class 3: Time-Sensitive
Description: Industry news, case studies, data-driven articles with statistics, trend analysis Expected freshness: 1-6 months Decay risk: High. Outdated statistics, superseded case studies, stale market data. Trigger for update: Any included statistic becomes more than 12 months old
Examples: "State of SEO 2025" reports, tool pricing articles, industry statistics roundups, vendor comparison posts
Class 4: Perishable
Description: Breaking news, event coverage, product launches, earnings reports, time-limited offers Expected freshness: Days to weeks Decay risk: Extreme. After the event, this content becomes historical and loses primary ranking value. Trigger for update: Typically converted to historical record or archived
Examples: Conference coverage, algorithm update news, new product announcements, quarterly earnings analysis
The 8 Decay Risk Elements
For any piece of content, audit each element type for freshness:
Element 1: Statistics & Data Points
Decay speed: Fast (6-12 months for most industries) Risk: The oldest statistic in an article becomes the content freshness anchor. If you cite a 2021 statistic in 2025 content, Google's systems detect the temporal mismatch.
Audit action:
- List every statistic with its source date
- Flag any statistic older than 18 months
- For competitive SERPs: check when competitor top-ranking articles last updated their stats
- Priority update: any stat that's cited prominently in the first 300 words
Element 2: Tool & Platform References
Decay speed: Moderate (12-24 months) Risk: References to deprecated tools, old UI screenshots, outdated feature names signal stale content.
Audit action:
- List every third-party tool mentioned
- Check if tool is still active, still named the same, and still relevant
- Google Analytics → GA4 transition is a prime example of this decay pattern
- WordPress Gutenberg vs. Classic Editor references date articles immediately
Element 3: Screenshots & UI References
Decay speed: Fast (6-18 months for SaaS interfaces) Risk: Screenshots showing old UI versions are a visual freshness signal. Users pogo-stick when screenshots don't match what they see.
Audit action:
- Catalog all screenshots in the article
- Compare against current UI of referenced products
- Priority: screenshots in step-by-step how-to content (highest user friction when outdated)
Element 4: API & Code Examples
Decay speed: Variable (tied to API versioning) Risk: Code examples using deprecated methods, removed parameters, or old syntax become misleading. High bounce rate for technical content with broken code.
Audit action:
- List every code block with the API/library version it applies to
- Check current documentation for breaking changes
- Add version annotations to code blocks going forward
Element 5: Regulatory & Legal References
Decay speed: Depends on legislative cycles Risk: Outdated compliance guidance (GDPR, CCPA, ADA, tax law) creates legal liability AND freshness decay signals.
Audit action:
- Identify all regulatory/legal claims
- Cross-reference with current regulatory status
- Add "last reviewed" dates to compliance-related content
Element 6: Competitor Comparisons
Decay speed: Fast (6-12 months in competitive markets) Risk: Product pricing, features, and market positioning change rapidly. Outdated comparisons hurt user trust and signal stale content.
Audit action:
- List all competitor names mentioned
- Check if pricing/features cited are still current
- "Best X alternatives" articles require quarterly review in active markets
Element 7: External Links
Decay speed: Continuous (link rot accumulates over time) Risk: Broken outbound links signal site neglect. The Historical Data patent monitors site maintenance patterns.
Audit action:
- Run all outbound links through a link checker (Screaming Frog custom extraction or Broken Link Checker)
- Replace or remove broken links
- High link rot (>5% broken) is a decay signal for older content
Element 8: Date-Anchored Claims
Decay speed: Immediate after the referenced period Risk: Phrases like "this year," "recently," "in 2022," "last quarter" create temporal anchors that make content feel stale as time passes.
Audit action:
- Search for: "this year", "recently", "last year", "currently", "as of [year]", "in [past year]"
- Replace time-relative language with specific dates or timeless phrasing
- "As of March 2025" is better than "recently" — it's honest and dateable
Decay Score Formula
For each page, calculate a Decay Risk Score (0-100, higher = more decayed):
Decay Risk = (Class Weight × 25) + (Element Risk Score × 75)
Class Weight:
Evergreen = 0.1
Semi-Evergreen = 0.3
Time-Sensitive = 0.6
Perishable = 1.0
Element Risk = average of 8 elements scored 0-1
(0 = fresh, 1 = fully decayed)Update Priority Matrix
| Decay Score | Action |
|---|---|
| 0-25 | Monitor quarterly |
| 26-50 | Schedule update within 90 days |
| 51-75 | Update within 30 days — ranking risk active |
| 76-100 | Emergency update — potential traffic loss underway |
The "Date Manipulation" Red Flag
The Historical Data patent specifically flags this behavior: changing the published or modified date on a page without substantively updating the content. Google can detect this via:
- Wayback Machine snapshots showing identical content with new timestamp
- No change in phrase co-occurrence or entity coverage
- No new inbound links coinciding with the claimed update
The rule: If you update a date, you must update the content. Touching only the metadata date is a manipulation signal.
What a Real Freshness Update Looks Like
A genuine freshness update that resets Google's decay clock:
- Replaces outdated statistics with current data (with sources)
- Updates screenshots to reflect current UI
- Adds a new section covering developments since last update
- Fixes or replaces broken outbound links
- Adds "Last updated: [date]" prominently in the content
A cosmetic update that does NOT reset the decay clock:
- Changing the published date only
- Minor typo fixes or style changes
- Adding a sentence to a section that was already current
- Changing headings without touching body content
Freshness Scoring Across Pages
For site-wide freshness monitoring:
- Export all pages with last-modified dates from your CMS
- Classify each page into the 4 content classes
- Calculate days since last substantive update vs. expected freshness window
- Flag any page where: days since update > (class freshness window × 1.5)
- Prioritize by traffic value × decay risk score