

Understanding The Impact of Google AI Overviews on Keyword Search Intent
In a previous article, we talked about the importance of keyword search intent. Google’s latest updates have highlighted the importance of quality content meeting users’ needs (this includes user queries). However, the introduction of Google’s AI overviews in May 2024 has left many bloggers, marketers, and business owners wondering what’s to come for SEO in the following years.
Google AI overviews are AI-generated summaries that answer user queries as quickly and directly as possible. Their objective is to improve the search experience and provide information in an easily digestible way, like bullet points and graphs.
As we already know, AI programs like ChatGPT and Google Gemini generate answers using information already on the web. In a study in September 2024, Rich Sanger determined that AI overviews show up in about 21% of searches, which means that not all types of keywords and inputs will generate an AI response.
This raises the question: What type of intent does Google consider most important for generating AI overviews, and how will this affect SEO strategies moving forward? To answer this question, I embarked on a deep dive into Google’s AI responses and what other SEO experts have to say about their effect on search engine optimization. I analyzed 40 keywords with different search intents to determine the rate at which these summaries appear in search results.
This article will explore Google AI overviews, how they work, expert opinions on how search intent affects these search results, and why optimizing them is essential for SEO in the coming years.
What are Google AI Overviews?
On May 14, 2024, Google announced the introduction of AI overviews in a slow rollout in the US. What had started as an experiment in Google search labs was now introduced to the general public at the top of sponsored and organic search engine results.
AI Overviews use Google Gemini’s language model to understand, reason, and break down complex user questions and provide simplified answers using data and information already on the web. For example, asking Google, “What are the top cities to spend the new year at?” will give you an AI-generated response like the one pictured below.


This overview gathered information and data from popular travel guides, transportation websites, and even other metasearch engines for travel like Skyscanner and Agoda. It provides suggestions in an easily readable format (bullet points) and links to other articles for further deep diving. Some SEO experts have described these results as “featured snippets on crack.”
These results have one objective: to improve a user’s search experience. The faster a user finds information online, the more satisfied they will be. The model has flaws, of course — like saying taking a bath with a toaster has health benefits — but it’s essential to understand how these AI-generated responses determine what the user is searching for when providing accurate responses.
So…
How do Google AI Overviews Work?
There’s no better way to begin understanding them than the source itself. Liz Reid, the VP and head of Search at Google, says these overviews are “… made possible by a new Gemini model customized for Google Search. It brings together Gemini’s advanced capabilities — including multi-step reasoning, planning, and multimodality — with our best-in-class Search systems.”
What exactly does that mean, though? It dates back to Google’s advanced language models like BERT and RankBrain. RankBrain, especially, was designed in 2015 to understand the intent behind a user query more accurately. The results might not explicitly match the keyword word-for-word, but they will match the user’s intent or what they’re really looking for when searching a term.
In 2018 however, BERT (Bidirectional Encoder Representations from Transformers) was developed as another model to understand natural language and context. So instead of understanding a query word by word, BERT was developed to comprehend nuance and context. This is what Bill Slawski (algorithm expert and Author of SEO By The Sea) had to say about this language model on X (formerly Twitter):


In layman’s terms, BERT is a fancy tool that helps computers understand and work with human language. Think of it like a “language tutor” for computers. It can learn how words fit together and their meaning in different contexts by predicting what a user means.
These new changes to Google’s AI learning and language models made it easier for search engines to perform tasks like:
- Answering questions based on the text given.
- Rewriting unclear queries into something clearer and easier to understand.
- Figuring out the role of each word in a sentence (like who is doing what)
- Finding names and key information in the text.
- Look at word context in all directions, not just left to right.
Before AI Overviews, users relied heavily on search results alone and had to sift through multiple articles to find answers to complex questions. Google’s Gemini takes everything learned from these Natural Language Processing models (NLPs) to improve search accuracy.
Gemini examines a lot of written information (books, articles, case studies, and websites) to determine how words and phrases are connected. It’s like teaching the AI to spot patterns—what words usually go together, what they mean in different situations, and how they relate.
In short, Google’s AI Overviews use Gemini’s complex language capabilities to find nuanced answers faster for the user.
We’ll revisit search intent’s impact on search engine results to understand how it affects these results and how these keywords and AI Overviews align.
What is Search Intent?
Rehashing our previous article about this topic, search intent is defined as why a user will type a particular phrase, keyword, or query in a search engine. It’s the reasoning and purpose behind a search action.
Google’s language models aim to understand search intent and provide the most accurate results. It’s why you get local business results for terms like “Christmas shopping near me” and online shopping results for terms like “Christmas lights for sale.”


Results for “Christmas shopping near me” show local stores on Google’s local pack. Google assumes you’re looking to shop in nearby, physical stores.


Results for “Christmas lights for sale” show Google Shopping options from different online retailers. Google assumes you’re attempting to buy something online.
The first keyword has a local and commercial intent, while the second has a transactional intent.
There are four types of search intents (five if you count local terms in a different category). Keywords are classified as follows:
- Informational: A user is trying to find information on a specific topic or an answer to a question, such as “How old is Taylor Swift?”
- Commercial: A user wants to decide or find information about a product or service. For example, “Play Station 5 reviews.”
- Transactional: A user is ready to take action, such as searching for a place to buy a product, sign up for a newsletter, etc. For example, “Charli xcx tickets for sale.”
- Navigational: A user wants a specific website, section, or brand, such as “Ticketmaster login.”
You can learn more about search intent and SEO in our previous blog post.
However, the point of this article is to explain how AI overviews behave based on search intent and common user queries.
Let’s break it down below.
The Role of Search Intent and AI Overviews
I compiled a list of 10 keywords per search intent type for this article. Each keyword was categorized as informational, commercial, transactional, and navigational. Then, I typed each of these terms on Google to determine the rate at which AI overviews appear based on the intent behind a user query, what type of results Google Gemini would throw, and what website appeared per search query.
Yes, all queries were holiday-related.
These were my findings:
Informational Intent & AI Overviews
Query | AI Overviews | Type of Results | Sources |
How to prepare for extreme cold weather | Yes | Bulleted, summarized how-to guide | Ready.gov, United Way, Weather.gov |
Types of extreme weather | Yes | Listicle with types of extreme weather and one-sentence summaries | National Oceanic Atmospheric Administration, World Meteorological Organization, Met Office |
Best way to layer clothing for cold weather | Yes | 3-step, bulleted listicle with short explanations | Heat Holders, Patagonia, AP News, SKRE |
Cold weather memes | No | Organic image results and meme websites | Pinterest, Tenor, Memebase, Instagram, Reddit |
Winter weather forecast | No | Organic results for forecast websites and recent weather news in “top stories” | Accuweather, National Weather Service, New York Times |
Winter Activities for Preschoolers | No | Rich snippet results from across the web | Pinterest, Lillio |
Best temperature to paint exteriors | Yes | Bulleted list summary from local businesses and paint stores | Improovy, Hirshfield’s, Diamond Star Painting, Groovy Hues, PaintRite Pros, Five Star Painting |
Benefits of winter boots | Yes | Short listicle with a brief explanation of their benefits | JK Boots, Hangwang Stories, Grisport, Foot Print Wellness |
How to wear winter boots with jeans | Yes | Mini-summary comparing different types of jean cuts and how to wear boots with them | The Mom Edit, YouTube Videos, Busbee Style, Erica Ball Style, Tecovas, Dream Pairs |
why do my ears hurt in cold weather | Yes | A bulleted list summarizing the possible reasons why ears hurt in cold weather | Harley Street ENT Clinic, Family Urgent Care, Miracle-Ear, Hearing Solutions, Columbus Speech & Hearing |
Out of the ten queries analyzed, 7 showed AI Overviews, meaning that potentially 70% of informational queries aim to be answered by Google’s artificial intelligence models.
Commercial Intent & AI Overviews
Query | AI Overviews | Type of Results | Sources |
Best jackets for winter | No | Comparison blog posts, Reddit results, and People Also Ask snippets | N/A |
Dr marten vs timberland boots | No | Comparison blog posts and People Also Ask snippets | N/A |
Top outdoor heater brands | No | Blog posts and articles comparing different products | N/A |
Columbia waterproof jacket alternatives | No | Rich snippets listing brands from across the web and organic results for comparison blogs | N/A |
Restaurants near me open during Christmas | No | Google local 3-pack and Google map results | N/A |
Snow removal services in Boston | No | Local pack and organic results for local businesses and directories like Yelp | N/A |
Winter tire comparisons | Yes | Listicles of what to consider when buying winter tires and brand benefits | Les Schwab, Fasttrack Emarat, Nokian Tyres, Black Circles Canada |
Polaris vs. Arctic cat prices | Yes | Brief, non-specific overview comparing ballpark prices of Polaris and Arctic cat vehicles | Composite Snowmobile Tracks, Nelson’s Speed Shop, CNBC, Octane |
Peppermint Mocha vs. Caramel Bruleé Latte | Yes | Overview comparing the characteristics of the holiday drinks from Starbucks | Starbucks, Yahoo News, The Every Girl,
USA Today, Business Insider |
Compare Starbucks coffee flavors | No | Articles directly from Starbucks and other sources to help users find the right coffee blend | Starbucks, Testing table, Reddit |
Three out of the ten commercial keywords analyzed turned up Google AI Overviews, meaning that around 30% of commercial queries could be answered by AI.
Transactional Intent & AI Overviews
Query | AI Overviews | Type of Results | Sources |
Buy winter boots | No | Local 3-Pack, Google Shopping results, online stores, and People Also Ask Snippets | N/A |
Winter jackets for sale | No | Online shopping websites from local stores, Local 3-Pack | N/A |
Winter safety checklist pdf download | No | Organic results from reputable sources with downloads | N/A |
Weather newsletter signup | No | Signup organic links from the weather channel, CNN, and NWS | N/A |
Christmas decorations for sale | No | Local 3-pack and organic results for Target and other Christmas shopping sites | N/A |
Christmas lights discounts | No | Online stores like Walmart with discounts and People Also Ask snippets | N/A |
Cheap winter decorations | No | Online stores like Amazon and Walmart that promote winter decor and Google Shopping products | N/A |
Purchase thermal jackets online | No | Organic results for online shopping | N/A |
Corduroy pants | No | Google Shopping results | N/A |
Outdoor space heaters | No | Google Shopping results | N/A |
None of these queries turned up Google AI overviews, meaning that Google still prioritizes Google Sponsored shopping ads and organic results for online stores for transactional keywords.
Navigational Intent & Google AI Overviews
Query | AI Overviews | Type of Results | Sources |
Reddit login | No | Organic results for Reddit’s login page | N/A |
Top floor marketing contact | No | Rich snippet with the direct phone number to click and call our offices, our Google Business Profile, and our website | N/A |
Dr marten boots for sale | No | Dr Marten’s website and local stores in the Miami-Dade County area | N/A |
winter deals amazon | No | Organic link to Amazon winter deals and People Also Ask snippets | N/A |
Best buy discounts | No | Organic results for Best Buy’s discount page and coupons | N/A |
Facebook Marketplace | No | Direct link to the Facebook marketplace in organic search | N/A |
Semrush academy | No | Organic, direct link to Semrush Academy | N/A |
TikTok trending videos | No | Organic links to trending topics on TikTok | N/A |
Holiday Marketing Tips Forbes | No | Marketing tip articles directly from the Forbes website | N/A |
No | Instagram organic link with rich snippets | N/A |
None of these navigational queries resulted in an AI overview.
Explaining The Data
Here’s what the data from 40 categorized keywords showed:
- Around 70% of informational keywords will yield an AI overview, while only 30% of commercial keywords will generate a response.
- Transactional and navigational keywords still rely on paid ads and organic search to yield results.
- AI Overview sources come from high-ranking websites with high authority. Informational and Commercial queries can pull results from news articles, online stores, topic-related websites, and local businesses.
- 25% of all researched keywords showed AI results, close to Rich Sanger’s research of 21% of search queries showing AI Overviews.
70%
Of informational keywords show AI Overviews
30%
Of commercial keywords show AI Overviews
0%
Of transactional and navigational terms show AI Overviews
25%
Of all researched keywords showed AI Overviews
What This Means for Keyword Intent & SEO
Google’s AI model will identify the user’s intent and produce an AI overview if it considers the query complex enough to warrant a quick summary or knows the user will search for multiple sources to feed their needs. It clearly understands nuance, like synonyms and context, but it’s more likely to show an AI overview if the query is complex.
This is what Juan Pablo Mendez Trongé, founder of Mauna Growth had to say about the future of SEO and AI Overviews:
“Honestly, Google’s AI overviews are a mixed bag, and by 2025, I think we’ll still be in a bit of a tug-of-war. On one hand, they make information easier for users to access, but on the other, they’re definitely eating into traffic that would’ve gone to publishers. It feels like Google-and other platforms-are starting to realize this isn’t sustainable. They need publishers to keep creating high-quality content, just like publishers rely on Google to drive traffic.


Juan Pablo Mendez Trongé founder of Mauna Growth
Other experts believe this is a good thing for the future of search engine optimization. According to Inge Von Aulock, Founder & COO of Penfriend:
“The key insight many are missing is that these AI Overviews need source material; they’re not creating content from thin air. This makes it more crucial than ever to be the authoritative voice in your space. If you’re not creating content AI can reference about your brand or industry, someone else will shape that narrative for you”.


Inge Von Aulock founder and COO of Penfriend
Personally, I don’t know what the future holds for Google’s AI and how query understanding and search engines will work in the coming years, but it’s clear that AI is going nowhere. Google will still search for sources that produce trustworthy content for AI overviews.
Producing articles and content that cater to a good user experience and answer users’ questions will still be the top priority for SEOs, and Dillon Loomis, SEO Consultant at Pinnacle Pursuit SEO agrees:
“Google’s AI overviews are changing how SEO works by focusing more on matching what people really want when they search. This makes it even more important to understand your audience on a deeper level. While AI is great at analyzing data, it doesn’t understand the feelings or needs behind a search. This is where SEOs can step up by creating helpful, problem-solving content that connects with users”.


Dillon Loomis, SEO consultant at Pinnacle Pursuit SEO
As an SEO expert, I can draw my own conclusions based on theirs and other expert’s opinions and my findings:
- Click-Through-Rates will drop. When users find the information they’re looking for at the top of the search results, there will be a diminished need to research further.
- Local businesses can still appear in AI Overviews, as seen in the query “Best temperature to paint exteriors,” so leveraging local SEO with well-written, informational content that stands out is key.
- Local, transactional, and navigational keywords don’t rely heavily on AI Overviews. E-commerce websites will only be impacted by more complex search queries where AI can throw results directly from Google Shopping.
- Google’s language models intend to simplify articles to make them easier for users to understand. Writing simple blog posts with colloquial or easy terms will increase their chances of appearing in AI overviews.
- Matching keywords to user intent is more important than ever. The pillar of content creation should be using more complex and long-tail keywords to target nuanced subjects.
- Building a brand and using other online marketing methods, such as social media and PR, will help businesses stand out in the competitive landscape of organic search.
Conclusion
Google AI overviews are the future of search results. SEO is constantly evolving, with Google’s constant updates and new approaches to search queries. It’s important to stay competitive in the landscape, and the rise of these AI-generated summaries emphasizes the importance of writing quality content focused on user intent.
While informational queries dominate AI Overview appearances, local, transactional, and navigational keywords still thrive in organic search. Redefining SEO strategies to adapt to these changes is crucial for 2025 and beyond. Going forward, incorporating long-tail and nuanced keywords to cater to different search intents is a strategy to consider.
All things considered, if you need a trustworthy partner for SEO, Top Floor Marketing is here to help you stay ahead in the changing world of SEO with AI. We’ll use our skills to understand what people search for, create great content, and improve local visibility to get your business noticed and drive results.
Reach out to us today to grow your online presence and stand out in this new AI-driven era!