A curated reference library for performance marketing and AI search. Each finding includes its primary source, scope and practical limits. These are third-party references, not original MyLeadsFactory research.
18% of the Google searches in Pew Research Center’s study produced an AI summary. The study used March 2025 browsing activity from 900 US adults and collected corresponding search results in April.
How to interpret it: This is a study-period finding, not the current share of all Google searches or a measure of MyLeadsFactory’s audience.
Pew observed a click on a traditional result in 8% of visits with an AI summary, compared with 15% of visits without one.
How to interpret it: The unit is visits, not the percentage of links clicked. This comparison does not establish what caused an individual website’s traffic decline.
The GEO paper reports improvements of up to 40% in its generative-engine visibility evaluations, with effectiveness varying across domains.
How to interpret it: An experimental visibility result is not a promised increase in live ChatGPT citations, organic traffic or leads.
Aggarwal et al., GEO: Generative Engine Optimization, 2024 revision
Google says its existing SEO practices apply to AI Overviews and AI Mode. Supporting pages must be indexed and eligible to appear with a snippet; no special AI file or schema is required.
How to interpret it: Eligibility does not guarantee inclusion. Do not sell a special markup format as a way to secure an AI recommendation.
Google says Ad Strength is not used to calculate Ad Rank or Quality Score, and does not directly determine serving eligibility.
How to interpret it: Review relevance, policy status and conversion outcomes separately. A higher Ad Strength label is not proof of better business performance.
Google describes the reported 1–10 Quality Score as a diagnostic tool and says it is not an input in the ad auction.
How to interpret it: Do not translate a one-point score change into a guaranteed percentage reduction in CPC. Investigate the underlying relevance and landing-page experience.
Google documents different conversion requirements for different campaign types when using Target ROAS.
How to interpret it: Check the applicable campaign and current documentation. A blanket minimum of 50 conversions is not a universal rule for every account.
Core Web Vitals assess loading, responsiveness and visual stability through LCP, INP and CLS.
How to interpret it: Separate field data from lab tests, record the sample and device group, and investigate poor experiences. Good scores alone do not guarantee high search rankings.
Last reviewed on 3 October 2026. Research dates describe the original source, not a new measurement of the market. Platform documentation can change. We removed claims whose linked sources did not support their precision or scope; absence from this library does not establish the opposite claim.
If a reference needs correction, contact info@myleadsfactory.com with the source and the specific issue. For practical application, start with our Google Ads audit checklist or review management pricing.
Tell us about your account. We reply within 24 hours to confirm fit and arrange read-only access; the review follows once scope and access are agreed.