What are AI Overviews?
AI Overviews (also referred to as AIO!) are a new type of search result from Google that use generative AI to answer the searcher’s query. These search results provide information relating directly to the query, citing websites who’s content has contributed to the answer provided.
This feature was originally deployed under the term, ‘Search Generative Experience’, or SGE. In May 2024, this was launched publicly as AIO within the US before being rolled out to 6 further countries: United Kingdom, India, Japan, Indonesia, Mexico and Brazil (correct at the time of writing!)
Which Searches Trigger an AI Overview Result?
So, how do you find out which queries will trigger an AIO result?
Whilst data is continuously changing as AI Overviews develops, early research by Zapier shows the below takeaways:
- Some of the top keyword topics that trigger AIOs include relationships (40.64%), food and beverage (23.58%) and business (18.48%).
- Topics that don’t tend to trigger AIOs as often include automotive (0.24%), education (3.44%) and travel (1.32%).
- Businesses that fall under the YMYL (your money or your life) category, such as finance, healthcare and insurance may find it more difficult to appear. This is due to the content being more influential on people’s lives, and as a result Google tends to feature articles from high-ranking authoritative sources.
AIO Content Sources
Growth Memo undertook an analysis of +500 AI overviews, to understand more about the feature.
In recent months, Google has been doubling down on ensuring that ‘helpful, reliable, people-first content’ sources are the ones appearing towards the top of search results. People-first content is as it says: content than is created primarily for people, not to manipulate search engine rankings.
Interestingly, Growth Memo discovered that websites such as Reddit and Quora – those providing more user generated and opinion-based content and usually highly favoured by Google – didn’t feature as highly within search results. This may be a response to some of the earlier testing results, where it was discovered that AIO brought up false information and harmful advice, typically from these types of websites.
In addition to this, they found that 59.6% of AIO citations were not appearing within the top 20 search results.
So ,what can we take from this?
Featured Snippets vs AI Overviews
“Only 6% of AIOs contain the search query, indicating user intent > exact match”
The above reaffirms that AIO doesn’t work exactly like featured snippets.
Featured snippet optimisation is ‘exact match driven’. This means you need to match the question and clearly indicate that the answer relates to the question in your content, for it to appear as a featured snippet.
In contrast, AI Overviews focus more on user intent and reflecting information that is more ‘useful’. This means the search query context is much more important than the exact wording.
In effect, we see AIO prioritising website credibility over SEO best practise.
Checking Domain Quality
Domain quality refers to the credibility or authority of a website within the context of an industry or topic, and is one of the Google ranking factors that affect how high a page ranks on search engine results pages.
As we’re all aware, strong SEO leads to improved ranking within search engines, but what else can help when it comes to domain quality?
- User Experience: Are users finding what they’re looking for when they land on your webpage? Is everything well organised and easy to understand or find?
- Content Quality: Is there information for users to take in, and does it make sense to what they are searching for? This relates to Google’s ‘want’ for helpful and high-quality content.
- Technical Optimisation: Is your website optimised from the backend, as well as the front? Things such as pages not loading, broken links, duplicate content and more can affect this.
AI Overviews Impact on SEO; is it still worth the time and effort?
In short, yes. We have seen a multitude of changes to the AIO functionality within its very limited lifespan so far, so can only assume it will continue to change and evolve. With this, SEO remains the constant, but there are still ways we can both Search Engine Optimise and AIO Optimise.
How can you target AI Overviews?
Explore searches using keywords that are relevant to your brand or company and see what appears in the AIO result. Clicking through each link that appears will show you a highlighted section of the website, indicating the exact wording that was used for the AIO citation.
From here, simply tweak your content to match the current answer, or look to provide an improved answer. This could be through changing the wording, or even changing the formatting of your content. For example, whether the information is laid out within a paragraph, as bullet points or within a table.
Reflecting this within your own content helps to improve the chances of ranking.
One of the challenges with AIO is being able to keep track of what information is appearing within searches. It means we must adjust our content accordingly, using the method above to consistently track searches.
Will AIO make my traffic drop?
There is some evidence that your traffic may see a drop if being cited in AIO. This could be due to the fact that AIO shows a much bigger extract compared to featured snippets, thus users are getting the answer they need direct from the search result.
Are AI Overviews Accurate?
AI Overviews use a custom version of the Gemini Language Learning Model (LLM) that incorporates Google’s Search Systems, meaning they work differently from chatbots and other AI tools such as ChatGPT, Poe etc.
Rather than just using training data to generate an answer AIOs are also using Google’s index and identifying the best, most relevant results for the query. So whilst it may not hallucinate, it may misinterpret questions and provide inaccurate answers.
Research by Neil Patel showed that inaccurate answers were the most common type of AIO ‘fail’, with 67.35% of respondents encountering it. Other issues included outdated answers and inappropriate content.
That’s not to say that AIOs are completely inaccurate though – as with most online content, it just takes a bit of common sense to look at the answer given and check that it is relevant, or from a quality source.
Lessons from this show that whilst AIOs can offer comprehensive answers, they should be fully relied on due to the limitations that exist around the LLM. These results prioritise content that is relevant and useful to the user, emphasising quality and authority.
Understanding how AIOs work and adapting your content strategy accordingly, can help to increase your chances of being featured in these search results. The strong prominence of social networks such as YouTube and LinkedIn appeared in citations shows there is an opportunity to impact AIOs with social and video content.
Has your content appeared in AIOs yet?
Share this