Introduction to AI Overviews
Google’s Search Liaison, Danny Sullivan, has shared insights into how AI Overviews work, explaining the role of predictive summaries, grounding links, and the query fan-out technique in shaping AI-generated search results. This information helps clarify why Google may link to websites that don’t match typical search results, while emphasizing that the fundamentals of search optimization remain unchanged.
Understanding AI Overviews
According to Danny Sullivan, the core principles of search optimization haven’t changed, even with the introduction of AI Overviews. He stated, "The core fundamental things haven’t really changed. If you’re doing things that are making you successful on search, those sorts of things should transfer into some of the things that you see in the generative AI kind of summaries." This means that websites that are already optimized for search should still perform well in AI Overviews.
Why AI Overviews Are Different
One of the main reasons AI Overviews are different from regular search results is the use of predictive summaries and grounding links. Predictive summaries provide answers to a search query while also trying to predict related variations of what a user might want to see. This is similar to Google’s Information Gain patent, which aims to predict the next question a user might ask after reading the answer to their current question.
Predictive Summaries
Predictive summaries are a key component of AI Overviews. They show answers to a search query and try to predict related variations of what a user will want to see. Danny Sullivan explained, "One thing I think that people find really confusing sometimes is that they’ll do a query and especially you’ll see …these are the top 10 results, but I don’t see them in the AIO, what’s going on? And it’s like, yeah, the query in the search box is the same query, but the model that’s going out there to try to understand what to show is kind of an overview, going beyond just the top 10 results."
Grounding Links
Grounding links are another reason why AI Overviews are different from regular search results. They are used to anchor the summary to verifiable information from the web ecosystem, providing a way for users to click and read about subtopics discussed in the answer summary. Danny Sullivan explained, "And then on top of that, it’s then also trying to bring in the grounding links. And those grounding links, because it kind of comes from a broader set aren’t just going to match. The queries are going to be different and the overall set is going to be different."
Query Fan-Out Technique
The query fan-out technique is used in AI Mode, which is currently in Google Labs. This technique involves using multiple related queries to understand the variations and related topics of a search query. Danny Sullivan explained, "…one of the things they talk about is like ‘we use an advanced query fan out technique with multiple related queries in it…’ And it’s basically that what I said before. You issued a query. You try to understand the variations and things that are related."
Don’t Mess Up Your Rankings
Sullivan cautioned against trying to rank for both organic and AI Overviews, as this can "mess things up." Instead, focus on optimizing for the core principles of search optimization, as these will still be relevant for ranking in AI Overviews.
Conclusion
In conclusion, AI Overviews are a new way of presenting search results, using predictive summaries, grounding links, and the query fan-out technique to provide a more comprehensive answer to a search query. While the links in AI Overviews may differ from those in regular search results, the core principles of search optimization remain unchanged. By understanding how AI Overviews work and focusing on optimizing for these core principles, websites can improve their visibility and ranking in search results. The three key concepts to understand are predictive summaries, grounding links, and the query fan-out technique, which work together to provide a more comprehensive and accurate answer to a search query.