SERP Volatility in the Age of AI

A comprehensive study by Authoritas, analyzing 11,203 keywords across multiple observation windows in 2024-2025, has revealed crucial data about how search results are evolving in the AI era. The research spans two key observation periods: an 8-week window (August 23 to October 17, 2024) and a 13-week window (October 17, 2024 to January 17, 2025), providing detailed insights into SERP volatility patterns.

Key Volatility Metrics

The study measured four critical types of volatility, with striking results across both observation windows:

First Window (~8 Weeks):

• SERP Volatility: 0.30
• Organic Rankings Volatility: 0.49
• AI Overview Rankings Volatility: 0.68
• AI Overview Text Snippet Volatility: 0.18

Second Window (~13 Weeks):

• SERP Volatility: 0.29
• Organic Rankings Volatility: 0.55
• AI Overview Rankings Volatility: 0.73
• AI Overview Text Snippet Volatility: 0.21

AI Overview Prevalence and Impact

The research revealed that AI Overviews appeared in only 18.8% of searches (2,104 keywords out of 11,203). However, their impact on these SERPs was significant. A critical finding showed that approximately 40% of top 10 organic ranking pages were not cited in AI Overviews, indicating a substantial disconnect between traditional rankings and AI-selected sources.

Correlation Analysis Findings

The study's correlation analysis revealed surprisingly weak relationships between different types of SERP changes:

• Organic Volatility & SERP Volatility: 0.19 correlation
• Organic & Generative Volatility: 0.09 correlation
• SERP Volatility & Generative Volatility: 0.04 correlation
• Generative Text Volatility & Generative Volatility: 0.13 correlation

These low correlations suggest that each type of change operates largely independently, requiring distinct optimization strategies.

Industry-Specific Volatility Patterns

The research uncovered significant variations in volatility across different sectors. The Beauty and Style and Travel categories showed notably higher organic ranking volatility compared to other industries. However, AI Overview volatility maintained consistently higher levels across all sectors, with volatility scores typically 20-30% higher than corresponding organic volatility scores.

Practical Examples from the Data

The study provided detailed case studies of high-volatility queries. For example, the query "how to repair scratched wood floor" showed dramatic SERP changes between observation periods, while "loan a kindle book to a friend" maintained remarkable stability. Similarly, for "chain management salary," nine out of ten organic ranking positions changed over the study period, exemplifying high organic volatility.

AI Overview Text Evolution

The research tracked specific changes in AI-generated content, finding that text snippets evolved significantly even when rankings remained stable. For instance, the query "What's the best way to store onions" showed three distinct versions of AI Overview text across the observation periods, each progressively moving from comprehensive storage methods toward more practical, immediate solutions.

Technical Implementation Data

The study evaluated SERP features based on their importance level, assigning specific values:

• Highest importance (1.0): Generative and generative trigger features
• Very high importance (0.9): Featured snippets
• High importance (0.8): Knowledge graph
• Medium importance (0.7): Shopping, news, video features
• Lower importance (0.5-0.6): Various supplementary features

Temporal Patterns

The data showed consistent patterns across time periods, with AI Overview volatility increasing slightly in the longer observation window (from 0.68 to 0.73), suggesting growing algorithmic refinement. This trend was accompanied by a similar increase in organic volatility (0.49 to 0.55), though at a lower absolute level.

Looking Forward: Data-Driven Predictions

The research suggests several quantifiable trends:

• Continued high AI Overview volatility (>0.70)
• Steady increase in organic volatility
• Persistent low correlation between different volatility types
• Growing gap between AI Overview and organic ranking patterns

These findings paint a picture of an increasingly complex search landscape where success requires sophisticated monitoring and rapid adaptation to multiple, independent ranking systems. The data clearly shows that traditional SEO metrics alone are no longer sufficient for understanding and optimizing search performance.

For SEO professionals, these numbers indicate the need for comprehensive tracking systems that monitor not just ranking changes but also AI Overview presence and content evolution. The low correlations between different types of volatility suggest that separate optimization strategies are needed for traditional and AI-driven results.