seoClarity introduced a new methodology for measuring artificial intelligence (AI) search performance on June 3, coinciding with Google’s launch of dedicated Search Console reports for AI Overviews and AI Mode.
The developments provide search engine optimization (SEO) professionals with enhanced tools and first-party data to understand and influence how content appears in AI-driven search results across platforms like Google, ChatGPT, Claude, Perplexity, and Gemini.
The seoClarity methodology employs a ‘golden set’ of prompts designed to cover the entire AI search funnel, categorized into strategic testing tiers, according to Suraj Lalchandani, seoClarity’s CEO.
Mark Traphagen, Vice President of Training and Methodology at seoClarity, said that since large language models (LLMs) do not permit traditional A/B testing, their approach utilizes a control group of correlated pages. This helps to isolate the impact of specific changes by filtering out noise from broader model updates or algorithmic shifts.
Mihir Naik, seoClarity’s Chief Product Officer, stated that the methodology emphasizes specific baseline periods and minimum test windows. He noted that measuring AI search responses requires sufficient time, as changes do not manifest immediately.
One key finding from seoClarity’s testing demonstrated causation in AI citations. Adding FAQ sections to test pages led to an increase in AI citations, while their subsequent removal caused citations to drop, the company reported.
AI authority is measured by stacking various signals, such as citation share on top prompts and consistency across different search engines, rather than relying on a single metric, according to seoClarity.
Google’s new Search Console reports, launched earlier this month, offer direct data on a site’s visibility within AI Overviews. This provides webmasters with unprecedented insights into how their content is being utilized and displayed by Google’s generative AI features.
Source: Search Engine Journal
