Contents

EN Underline
  1. AI Overviews Explained: Mastering the Basics
  2. How Google's AI Overviews Work
  3. AI Overviews vs SGE: Spot the Difference
  4. How Are Users Reacting to AI Overviews?
  5. Adapting Your Strategy for AI Overviews
  6. FAQs

Get Weekly Marketing Tips

Join 30,000+ marketers and get the best marketing tips every week in your inbox

AI Overviews are Google’s latest search feature that uses generative AI to summarise search results. Users get quick, concise answers to certain queries directly within the search engine results pages (SERPs), with links to relevant web pages.

But how does it work, and what does this mean for your search strategies?

AI Overviews Explained: Mastering the Basics

Here’s the basics of what you need to know:

  • What Are AI Overviews? AI Overviews are AI-generated summarised answers to queries, pulling data from Google’s Knowledge Graph or multiple sources across the web.
  • When Were AI Overviews Introduced? Google launched AI Overviews in May 2024 in the U.S., expanding to the UK and other countries in August 2024.
  • Where Do AI Overviews Feature? AI Overviews appear at the top of Google’s search results, above organic listings, Featured Snippets, and paid ads.
  • Why Do AI Overviews Matter? AI Overviews shift search behaviour by increasing zero-click searches, meaning businesses must optimise content to be referenced rather than relying solely on traditional rankings.
  • Who Benefits? Users get faster answers, while businesses featured in AI Overviews gain brand exposure and credibility despite potential traffic declines.

Struggling to know how to use AI with your marketing?

Download a free copy of AI Marketing: Everything You Need to Know and get immediately up-to-speed

Download my free copy
Front cover of Exposure Ninja's "AI Marketing: Everything You Need to Know" guide.

How Google’s AI Overviews Work

Google’s AI Overviews use generative AI models, like PaLM 2. These models scan web pages, process data, and generate a summarised answer in seconds.

Here’s how the process works:

  1. Query Analysis: When a user enters a search, Google determines whether a generative response is best based on the intent and complexity of the query.
  2. Content Retrieval: Google’s AI scans its index of high-ranking web pages, news sites, and authoritative sources to gather relevant information.
  3. Information Processing: The AI extracts key insights from multiple sources, cross-referencing details to improve accuracy and avoid misinformation.
  4. Summary Generation: The AI generates a concise, structured response, often incorporating bullet points, lists, or step-by-step instructions for readability.
  5. Source Attribution: AI Overviews include links and link cards to sources, allowing users to explore further while ensuring credibility.

 

Screenshot of Google's AI Overview to the query "is SEO still relevant" showing a concise answer, a bullet list, and link cards to the websites it parsed for information

AI Overviews vs SGE: Spot the Difference

AI Overviews evolved from Google’s experimental Search Generative Experience (SGE). While SGE provided extensive AI-generated content, AI Overviews offers shorter, more focused answers.

Another major difference is interactivity. SGE allowed users to engage in a chat-like experience, where they could ask follow-up questions to refine results further. AI Overviews has removed the chat feature entirely, making AI Overviews feel more like an enhanced Featured Snippet than an AI chatbot.

AI Overviews also place more emphasis on source transparency by prominently displaying links to referenced websites. This change is beneficial, allowing websites producing credible, high-quality content the opportunity to rank higher in SERPs.

How Are Users Reacting to AI Overviews?

Google’s AI Overviews have had a slightly “rocky” rollout. Shortly after its U.S. debut, Google’s AI suggested people consume one small rock a day to increase their iron intake and eat non-toxic glue on pizza to make cheese stickier.

This shaky start seems to have impacted how people perceive AI Overviews. When performing keyword research, we noted many keyword variants related to “how to turn off AI Overviews”, indicating frustration among some users.

Screenshot of Semrush Keywords showing 11 variations of "how to turn off AI overviews" with search volumes ranging from 720 to 40

While some users are sceptical, Google is refining this feature based on user interactions, feedback, and evolving AI training models. By integrating AI Overviews with its core ranking systems, it prioritises accuracy by pulling from high-quality, authoritative sources.

This presents an opportunity for businesses. If your content is well-structured, reliable, and optimised for AI-driven search, you stand a better chance of ranking in AI Overviews. This makes AI optimisation essential to modern search engine optimisation (SEO) strategies.

Adapting Your Strategy for AI Overviews

To stay visible in AI-powered search results, digital marketing strategies must incorporate Generative Engine Optimisation (GEO). This means:

  • Publishing accurate, high-authority content.
  • Structuring content with clear headings and concise answers to increase the chances of being referenced in AI Overviews.
  • Building credibility through backlinks.

Want to see how your website stacks up against GEO best practices? Request a free website and marketing review today.

To learn more about GEO for AI Overviews, read our latest blogs:

Or watch:

FAQs

Where does AI Overviews get its information?

Google’s AI Overviews pull data from a mix of high-ranking web pages, news sites, authoritative sources, and Google’s Knowledge Graph.

Why did Google add AI Overviews?

Google’s AI Overviews are in response to the rise of generative AI. AI Overviews aim to speed up search results and enhance user experience by providing quick, AI-generated answers to user queries.

Has Google rolled back its AI Overviews search feature?

Not currently. Despite criticism, Google is still refining AI Overviews. It may make changes or offer an opt-out feature in the future.