AI & TechArtificial IntelligenceBigTech CompaniesDigital MarketingNewswireTechnology

Google AI Overview Shows Different Data for Commercial Searches

▼ Summary

– Peec AI found that Google’s AI Overviews appeared in about 87% of 500,000 prompts it studied, which skews toward commercial buying-intent queries.
– For decision-stage prompts like product comparisons, the rate was 88.5%, with longer prompts triggering AI Overviews more often than shorter ones.
– The 87% figure is much higher than broad studies like Ahrefs (20.5%) because Peec’s dataset excludes navigational searches and focuses on high-intent queries.
– Within the EU, AI Overviews appeared in 76% of sampled searches, versus 90.3% outside it, with France at 0% since the feature hasn’t launched there.
– Peec used machine-learning models to sort prompts by funnel stage and measure AI Overview presence, but hasn’t published accuracy figures for those classifiers.

A new study from Peec AI, an AI-powered search visibility platform, reveals that Google’s AI Overviews appear in roughly 87% of the 500,000 search prompts it analyzed. This figure, however, is not representative of all Google searches.

The dataset is deliberately skewed toward commercial, buying-intent queries, excluding navigational searches like brand-only terms. As a result, the 87% rate describes a specific, high-value slice of search behavior, not the broader search landscape.

What Peec Discovered

For decision-stage prompts , the kind users type when comparing products , the AI Overview appearance rate hit 88.5%. Longer queries consistently triggered AI Overviews more often. Two-word searches saw a 64.6% rate, while prompts of 11 to 15 words peaked near 89%.

Regional differences also emerged. Within the European Union, AI Overviews appeared in 76% of sampled searches, compared to 90.3% outside it. France was an outlier at 0%, as Google has not yet launched AI Overviews or AI Mode there.

Why the Number Is So High

The 87% figure is significantly higher than what broader keyword studies report. For context, Ahrefs analyzed 146 million results and found AI Overviews on 20.5% of keywords. A randomized field experiment we covered last month saw them on 42% of queries. Earlier samples from Authoritas and SE Ranking landed near 30%.

The discrepancy boils down to what Peec measured. Its dataset consists of business-oriented prompts, mostly commercial research and product comparisons, and excludes navigational searches like standalone brand names. In Peec’s sample, those high-intent, fuller-sentence prompts triggered AI Overviews far more often. Remove the navigational and short-tail queries that rarely generate an overview, and the average climbs.

So the 87% figure measures how often AI Overviews appear for buying-intent prompts, not how often they appear across all of Google.

How the Test Worked

Peec drew its prompts from its own dataset, sampled in April. It used machine-learning models to sort prompts by funnel stage and query type, then measured how often AI Overviews showed up for each. Peec has not published accuracy figures for those classifiers, so the labels are model-assigned rather than hand-checked.

Why This Matters

What truly matters for your visibility is where AI Overviews show up. In Peec’s sample, they appeared on most decision-stage prompts, like comparison and “best X for Y” searches that shoppers use when comparing options. In these cases, the AI Overview is often the first thing a buyer sees, so being featured here can directly influence which brands they consider.

If a visibility effort targets only informational searches, it misses the searches that lead to actual purchases.

Looking Ahead

AI Mode is the surface to watch next. It passed a billion monthly users in its first year, and Google is now linking it to AI Overviews so a follow-up question moves a user from the overview into AI Mode without leaving Search.

(Source: Search Engine Journal)

Topics

ai overviews prevalence 95% search query analysis 92% commercial buying intent 90% methodology limitations 88% visibility implications 87% regional variation 85% comparative studies 84% data sampling bias 83% ai mode growth 82% query length impact 80%