Ahrefs analyzed 146 million search results to determine which query types trigger AI Overviews. The research tracked AIO appearance across 86 keyword characteristics.
Here’s a concise look at the patterns and how they may affect your strategy.
What The Analysis Found
AI Overviews appear on 20.5% of all keywords. Specific query types show notable variance, with some categories hitting 60% trigger rates while others stay below 2%.
Patterns Observed Across Query Types
Single-word queries activate AIOs only 9.5% of the time, whereas queries with seven or more words trigger them 46.4%. This correlation indicates that Google primarily uses AIOs for complex informational searches rather than simple lookups.
The question format also shows a similar trend: question-based queries result in AIOs 57.9% of the time, while non-question queries have a much lower rate of 15.5%.
The most significant distinctions are seen based on intent. Informational queries make up 99.9% of all AIO appearances, while navigational queries trigger AIOs just 0.09%. Commercial queries account for 4.3%, and transactional queries for 2.1%.
Patterns Observed Across Industry Categories
Science queries have an AIO rate of 43.6%, while health queries are at 43.0%, and pets & animals reach 36.8%. People & society questions result in AIOs 35.3% of the time.
In contrast, commerce categories exhibit opposite trends. Shopping queries are associated with AIOs only 3.2% of the time, the lowest in the dataset. Real estate remains at 5.8%, sports at 14.8%, and news at 15.1%.
YMYL queries display unexpectedly high trigger rates. Medical YMYL searches trigger AI Overviews 44.1% of the time, financial YMYL hits 22.9%, and safety YMYL reaches 31.0%.
These findings contradict Google’s focus on expert content for topics that could impact health, financial security, or safety.
Queries With Low Presence Of AI Overviews
6.3% of “very newsy” keywords trigger AI Overviews, while 20.7% of non-news queries display AIOs.
The pattern indicates that Google deliberately limits AIOs for time-sensitive content where accuracy and freshness are essential.
Local searches demonstrate a similar trend, with only 7.9% of local queries showing AI Overviews compared to 22.8% for non-local queries.
NSFW content consistently avoids AIOs across categories: adult queries trigger AIOs 1.5% of the time, gambling 1.4%, and violence 7.7%. Drug-related queries have the highest NSFW trigger rate at 12.6%, yet this remains well below the baseline.
Brand vs. Non-Brand
Branded keywords show slight differences compared to non-branded ones. Non-branded queries trigger AIOs 24.9% of the time, whereas branded queries do so 13.1% of the time.
The data indicates that AIOs occur 1.9 times more frequently for generic searches than for brand-specific lookups.
No Correlation With CPC
CPC shows no meaningful correlation with AIO appearance. Keyword cost-per-click values don’t affect trigger rates across any price range tested, with rates hovering between 12.4% and 27.6% regardless of commercial value.
Why This Matters
Publishers focused on informational content encounter the greatest AIO exposure. Question-based and how-to guides align closely with Google’s trigger criteria, putting educational content publishers at the highest risk of traffic loss.
Medical content has the highest category-specific AIO rate, despite concerns about AI accuracy in health advice.
Ecommerce and news publishers are relatively less affected by AIOs. The low trigger rates for shopping and news queries indicate these sectors experience less AI-driven traffic disruption compared to informational sites.
Looking Ahead
Using this data, publishers can review their current keyword portfolios to identify AIO exposure patterns. The most reliable indicators are query intent and length, with industry category and question format also playing significant roles.
AIO exposure varies considerably across different industry categories, with differences exceeding 40 percentage points between the highest and lowest. Content strategies need to consider this variation at the category level instead of assuming consistent baseline risk across all topics.
For a more in-depth examination of this data, see the full analysis.
Featured Image: Zorion Art Production/Shutterstock