Google’s Liz Reid recently discussed what goes on behind the scenes of AI Search and introduces a relatively new concept, Browsy Queries. Her feedback offers insights on what SEOs should be focusing on right now in order to perform better in AI search surfaces.
Search Behavior Is Varied, Not Monolithic
Host Joe Wazenthal asked Liz Reid about user behavior patterns in search, how users choose to use classic search or AI search, and what differences in queries result from choosing one platform over the other.
Liz Reid answered by first defining what she is talking about, linking classic search and AI Mode together as Search, then positioning Gemini as something else that is fundamentally different.
She also stated that there are a massive amount of users whose search behaviors are varies across all search surfaces, in essence saying that there isn’t a monolithic user behavior pattern in which people are doing the exact same searches, the patterns the interviewer was looking for in his question.
Liz Reid answered:
There’s sort of your main search page. There’s AI Mode. That’s part of search.
And then there’s the Gemini app.
And I would say there’s a lot of users, so their behavior varies across all of them.”
Search And AI Usage Patterns Are Complex
The SEO and publishing community often thinks about Search as Google but Liz Reid says that user behavior patterns point to a more complex search ecosystem where users are relying on multiple platforms.
She continued her answer:
“But there are some patterns. There’s plenty of people who co-use across them. There’s plenty of people that are actually using several AI products right now, just in general, not even just within Google.
Across Gemini and Search, the more informational ones… Like, if it’s an informational query, then the probability that they’re using Search or AI Mode is going to be higher.
If it’s a creative query, it’s like more of a productivity question like, please rewrite this to make it sound more formal, right? Those type questions are going to be more Gemini-oriented.
Between AI Mode and Search, the main search page, some people use AI Mode mostly via AI overviews. They start in AI overviews and they transition.
For those who go direct to AI Mode, they tend to do that for queries that they consider sort of more complex, longer questions, questions where they expect that they’re going to do more follow-ups, versus if you’re doing a very browsy query, you might choose to prefer all of the SERP.”
Browsy Queries And Browse Search Intent
When we think about search, it may be useful to consider that people not only search across platforms, but they do it for different reasons.
Takeaways About How People Use AI
- Co-Users
People use multiple platforms simultaneously (co-use) - Informational Queries
These tend to happen on Classic Search and AI Mode - Creative Queries
These tend to happen on Gemini - AI Mode Direct
Queries that originate on AI Mode, where people navigate to AI Mode, tend to be complex, what was traditionally called longtail. - Browsy Queries
This is a relatively new phrase that Googlers apparently use.
What Are Browsy Queries?
The phrase “browsy queries” must be something that Googlers use internally and maybe is more familiar with people who do Pay Per Click advertising. There aren’t really many instances of the phrase but here’s how Google uses it.
A software engineer formerly of DeepMind and Google describes in her LinkedIn Profile having created a machine learning model that identifies “browse intention” queries on Google Search, an invention that improved click-through rates by 5%.
She wrote:
“Built a machine learning model to identify ‘browse intention’ query on Google Search, which presents engaging content on search result pages for browsy queries (e.g. “best places to visit in Orlando”). Improved global search result click-through rate by 5%”
The phrase “browsy queries” is also used in a Google job description for a commerce software engineer, placing the phrase in the context of shopping queries.
“Commerce Retrieval researches and develops high-precision algorithms to reduce the search space for product queries by 8 orders of magnitude under tight latency and compute constraints. Our solutions are tailored to the unique complexities of the Shopping domain including browsy queries, a hierarchical schema, and short multimodal documents.”
It’s also used in the context of video ads in a Google support page for video ads:
“These new shoppable formats will be shown to potential customers in lower intent, more “browsy” Search placements earlier in their shopping journey.”
What Browsy Queries Means And How To Optimize For it
What’s consistent across all three uses is that “browsy queries” are defined by a discovery-level intent stage.
In each example, Google is identifying what the user keep the user exploring:
- The DeepMind example ties browsy queries to engaging content that a user wishes to browse through, not direct answers.
- The commerce job role positions browsy queries as a quality of commerce search.
- The ads example places browsy queries earlier in the shopping journey at about the discovery phase.
The useful takeaway is that Google treats these queries as exploration problems. What makes browsy queries complex is that they have under-specified user intent and are the result of consumers who may be looking for inspiration.
For an SEO or an online merchant, it means that a user has intent but hasn’t narrowed down what they want. That’s where contexts like “Stylish Outfits For Summer” come in handy. Broad keyword phrases are probably useful here. I like a pyramid structure where the deeper a user gets into a page, the more specific it may become.
Browsy Search Patterns And Organic Search
Browsy queries are a search behavior that is different from direct answer-seeking. These are searches where users appear to want options, inspiration, comparisons, or a chance to keep exploring.
That distinction matters because AI Mode is not necessarily the best surface for every query. Reid’s comments suggest that some searches still benefit from the organic search rseults because the a synthesized answer is not what they need. Browsy query users may want to see choices, scan sources, compare possibilities, and follow their own path.
For SEOs, it’s a reminder that not every query should be treated as a direct-answer problem. Discovery-stage content may need to support exploration before narrowing the user toward a more specific answer, product, or decision.
Key points:
- Browsy queries appear to describe searches with discovery-level intent.
- These queries may favor the full SERP because users want to browse options.
- The phrase appears in Google contexts related to Search, shopping, and ads.
- Broad, inspiration-oriented content may still matter in AI Search.
- A useful page structure may start broad, then become more specific as the user moves deeper into the content.
Browsy queries are a reminder that search still includes discovery, and discovery may be best served when users can see multiple paths instead of one synthesized response.
Watch the Liz Reid interview here:
Google’s Liz Reid on Who Will Own Search in a World of AI
Featured Image by Shutterstock/Summit Art Creations
