How to Optimize Content for AI Overviews
To optimize content for AI overviews, use clear headings, concise bullet points, and rich media like images or videos. Implement schema markup (FAQ or Article) to help AI interpret the content. Focus on CSQAF: Citations, Statistics, Quotations, Fluency, and Authoritativeness. Answer critical questions early in the text and break down large paragraphs into scannable lists to improve readability and increase the chances of being featured in AI-driven results.
Here’s what we’ll cover:
- What Is the Fundamental Difference Between SEO and AEO?
- Understanding AI Overviews
- The Balancing Act: Google and Content Creators in the Era of AI Overviews
- How AI Overviews Differ from Traditional SEO
- When Are AI Overviews Being Triggered?
- AI Overviews and Their Relationship with Other SERP Features
- Why Is AEO a Fundamental Shift From SEO?
- How to Write and Optimize Content for AI Overviews
- Navigating the New Frontier of AI Overviews
What Is the Fundamental Difference Between SEO and AEO?
The main difference between AEO (Answer Engine Optimization) and SEO (Search Engine Optimization) is their focus. SEO traditionally targets improving rankings through keyword usage, backlinks, and metadata for search engines. AEO, on the other hand, optimizes content for AI systems by focusing on concise answers, authoritativeness, and strong factualness. AEO emphasizes generating direct, conversational responses that AI overviews can summarize, prioritizing user experience and conversationality over long-form, keyword-heavy content.
Understanding AI Overviews
AI Overviews are AI-generated summaries displayed at the top of Google search results. They provide quick, concise answers to user queries, often supported by links to relevant websites.
Historical Context of AI Overviews and SGE:
On 14 May 2024, Google introduced AI Overviews, essentially their take on enhanced search capabilities powered by artificial intelligence. It had been in a test phase for most of 2023, initially known as SGE or Search Generative Experiences. Still, that name was a real pain to retain, plus other very embarrassing incidents and daily iterations had them settle with the AI overviews. Currently, it’s only directly available to users in the United States, though individuals in other countries can still access it by using a VPN that makes it appear as if they’re in the US.
More importantly, AI overviews are constantly changing; what’s settled on today by the community and what Google says itself can all change tomorrow.
What do we know for sure about AI Overviews?
- AI overviews are rolling out before the end of 2024 worldwide
- They appear for informational queries more than commercial or transactional ones
The Balancing Act: Google and Content Creators in the Era of AI Overviews
Google’s success hinges on maintaining a delicate balance between the interests of webmasters, content creators, and its evolving technologies. This balancing act isn’t new—think back to when featured snippets were first introduced.
The symbiotic relationship between Google and websites is clear: Google needs content to fuel its algorithms, and websites rely on Google to drive traffic.
Google asserts, “With AI Overviews, people visit a greater diversity of websites to help with more complex questions. And we see that the links included in AI Overviews get more clicks than if the page had appeared as a traditional web listing for that query.” However, some in the industry are raising concerns.
Garett Sussman of IpullRank points out, “There’s always been this unspoken contract between Google and websites. You can index our content as long as you send us traffic. AI Overviews feels like it’s breaching that contract.”
How AI Overviews Differ from Traditional SEO
Traditional SEO focused on ranking at the top of the search engine results page (SERP). AI Overviews prioritize highly targeted, precise content that directly answers user queries. Even well-ranking traditional content might not appear in AI Overviews if it does not meet the specific criteria for relevance and clarity.
According to a study by BrightEdge, results appearing in the AI overviews are not ranking in the original results 66% of the time.
Essentially, what we are seeing today and in the future of search is the “complexification” of the queries, which mirrors our growing demand for a better knowledge experience. No one wants to comb through information adjacent to their core ask, and generative AI will further this desire.
When Are AI overviews Being Triggered?
- The prevalence of AI Overviews varies by industry.
- AI Overviews are more common in healthcare and B2B technology, where inquiries are more informational. If you’re in B2B tech marketing like myself, you will be affected by AI overviews, so start preparing for this dramatic shift.
- Industries such as restaurants or travel, which focus on transactional interactions (e.g., reservations, directions, calls), tend to have fewer AI Overviews.
A notable exception is the finance sector, where Google rarely deploys AI Overviews, possibly due to YMYL. However, there is currently no data to corroborate this.
AI Overviews and Their Relationship with Other SERP Features
According to a study by Advanced Web Ranking, which analyzed 8,000 keywords, AI Overviews often appear alongside other SERP features, such as People Also Ask, Videos, and Featured Snippets. The study found that optimizing content to appear in these SERP features increases the likelihood of being included in AI Overviews.
In nearly all instances (99.5%), the presence of AI Overviews is linked to including the People Also Ask SERP feature.
- People Also Ask: Appears in 99.5% of cases when AI Overviews are shown.
- Videos: Found in 69.7% of cases alongside AI Overviews.
- Featured Snippets: Displayed in 59.5% of cases.
- Discussions & Forums: Present in 58.8% of cases.
- Images: Included in 30.5% of cases.
This suggests that when AI Overviews are triggered, it’s often due to broader search intent, which activates related SERP features.
The study also found that when AI Overviews are absent, other features like Products, Local Packs, Sitelinks, and Knowledge Panels dominate the SERP.
What is the difference between AI overviews and featured snippets?
- AI overviews: Aggregate information from multiple sources to provide comprehensive responses.
- Featured snippets: Extract concise information from a single source to directly address specific queries.
Example of an AI overview for a closely related question
Why Is AEO a Fundamental Shift from SEO?
In a groundbreaking study titled GEO: Generative Engine Optimization, researchers from several prominent institutions studied the GEO (AEO) framework, which offers a new approach to optimizing content for Generative Engines (GEs).
These engines, like BingChat and our AI Overviews, use large language models (LLMs) to synthesize and present information directly to users, bypassing traditional SEO strategies.
Unlike traditional SEO, which relies heavily on keyword optimization and backlinks, AEO requires targeted, domain-specific content optimizations such as simplifying language, adding credible sources, and improving fluency. The transition to AEO reflects the shift in user preferences from a ranked list of web pages to AI-generated, synthesized answers, fundamentally altering how content creators must approach visibility and relevance.
The study evaluated nine tactics across 10,000 search queries to enhance visibility:
- Authoritative: Content was adjusted to make claims in a more persuasive, authoritative tone to increase trustworthiness.
- Keyword Stuffing: Keywords were added to mimic the same negative practice known in traditional SEO.
- Statistics Addition: Quantitative data replaced qualitative statements to add concrete evidence.
- Cite Sources: Credible citations were added to support claims and boost content legitimacy.
- Quotation Addition: Quoting reliable sources was used to enhance authority and context.
- Easy to understand: Language was simplified to make content more accessible and comprehensible.
- Fluency Optimization: Content was edited to improve the natural flow and readability.
- Unique Words: Distinctive terms were incorporated to stand out in the generated results.
- Technical Terms: Domain-specific terminology was added to enhance relevance in niche queries.
Results showed a potential 40% boost in content visibility through AEO-specific methods.
One of the most significant findings was that lower-ranked websites in traditional search engine results (SERPs) gained disproportionately more visibility in GEs using the Cite Sources and Statistics Addition methods, leveling the playing field between small and larger, established ones. For instance, websites ranked lower in SERPs saw visibility improvements of up to 115.1%, while top-ranked sites experienced a 30.3% decline.
This study has far-reaching implications for content creators in the AEO universe however, like any study, it has a few limitations.
AEO Study limitations:
- Black-box nature of generative engines: The proprietary algorithms of generative engines make it difficult to fully understand how specific optimizations impact visibility, as creators have limited insight into how their content is processed.
- Reliance on synthetic and limited queries: The study used artificially generated queries and a limited set of 10,000 questions. This may not reflect the full range of real-world search behavior, reducing the generalizability of the results.
- Limited evaluation on real-world platforms: The study’s results were primarily tested on Perplexity.ai, which may not fully represent other widely used platforms like Google or Bing.
How to Write and Optimize Content for AI Overviews
Let’s start by saying that writing spammy content generated by AI, where someone asks a language model (LLM) to “write them an article on the benefits of B2B content marketing,” will not get you far. There are many risks of over-reliance on AI in content marketing,
Moreover, seasoned writers and content creators can now accurately and predictably spot AI writing, so don’t surrender and outsource all your brain power to the LLMs, please don’t.
Stephanie Trovato recently noted, “…this doesn’t mean you shouldn’t use AI to support your content creation. It just means you need to make sure the content is beneficial to readers,” echoing many of us in the community’s sentiments exactly.
Step 1: Understand the fundamentals of Generative Engine Optimization
Creating an article that performs well for search engines differs from crafting one for AI models.
SEO prioritizes:
- Keyword use
- Backlinks
- Overall searchability
AEO focuses on:
- Producing well-structured content (lists, breaking up long text, etc.)
- Abundant sourcing
- References
The uniting technicality between an article performing well in both universes is structured data. Structured data will continue to be relevant in AEO as it was in SEO, as Ryan Sylvestre notes:
“Structured data is no longer a nice to have when it comes to SEO and digital marketing. The release of AI search summaries from Google sends a clear message to marketers that content organization is a must.”
To enhance visibility and improve chances of appearing in featured snippets or knowledge graphs:
- Clear Structure: Use descriptive headings and subheadings to organize your content.
- Concise Information: Answer common questions directly with bullet points, lists, and short paragraphs.
- Rich Media: Embed images, videos, and diagrams to boost chances of appearing in rich snippets.
- Schema Markup: Apply relevant schema types, such as FAQ or Article, and ensure it’s validated with tools like Google’s Structured Data Testing Tool to improve search engine understanding and display.
Step 2: Use CSQAF: Citations, Statistics, Quotations, Fluency, and Authoritativeness to improve your content writing for AEO
These are the remaining factors from the original 9 we saw in the seminal study by Aggarwal et al. (2024) on GEO, but we removed the ones that are detrimental to improvement in AEO land.
Five factors improve any piece of content for generative engines:
- Citations: Including references from credible sources increases visibility.
- Statistics: Adding quantitative data strengthens content relevance.
- Quotations: Using direct quotes from authoritative sources enhances trust.
- Fluency and Readability: Optimizing language for clarity and ease of understanding boosts engagement.
- Authoritativeness: A persuasive and authoritative tone helps content stand out in generative engine results.
Examples of GEO methods show how small additions and deletions can significantly enhance content visibility. Additions are highlighted in green, and deletions are in red (Aggarwal et al. 2024)
According to the study, these strategies can lead to up to a 40% improvement in content visibility. That is a staggering number for content creators to consider.
Step 3: When writing content to appear in AI Overviews, answer the question early
This step comes from Exposure Ninja’s Tim Cameron-Kitchen. He mentions that placing H2 tags under the blog post’s main title, where the article’s central question is answered in “short sentences and straightforward language,” caused their article to rank in the AI overviews after only appearing in the organic results.
I particularly like his term “Exact Match Blogs” (EMBs), which refers to short content that clearly and swiftly answers specific questions. This approach contrasts with traditional SEO, where the content would be extensive and cover a primary keyword and related questions in the subheadings. AI integration and AEO focus on conversationality, with people seeking direct answers to their problems and questions. As a result, EMBs are likely to replace the previous approach.
Step 4: Break down monolithic paragraph structures
This one seems like a real pain to many writers, and it doesn’t mean we have to forfeit good prose for lists and bullet points. In GEO, the whole point is to help AI scan your articles quicker and better. Stephanie Trovato’s Hubspot article corroborates this step but calls it “emphasize scanability.”
If you write a lovely piece of prose to emphasize a narrative, style, or tone, do it, but provide a broken-down and to-the-point list of the same information above or below and use a logical connector to link it all. Remember to mark your subheadings, lists, bullet points, etc.
The chunk of writing:
A website’s performance in a user’s browser can be affected by the size of the HTML, the use of external resources, and the way CSS and JavaScript are utilized. Dr. Jatinder Manhas from the University of Jammu points out that making minor adjustments to the structure of HTML can result in significant reductions in file size. Minimizing JavaScript and opting for CSS-based solutions can significantly enhance speed. Implementing lazy loading for images ensures they’re only loaded when needed, cutting down on initial loading times. Additional enhancements for the front end include limiting custom fonts, employing SVGs for icons, and incorporating critical CSS directly. With strategically designed improvements, these methods can substantially increase the speed of your website.
Broken down into a list:
General Advice for Developers and Website Owners:
- HTML Optimization: Simplify your HTML structure to save on file size. Removing optional tags (e.g., </li>, </p>) can reduce the file size by 5-20%.
- Reduce JavaScript Usage: Minimizing JavaScript and relying on CSS for styling can drastically improve performance.
- Image Optimization: Use optimized image formats like WebP and employ lazy-loading techniques to improve load times.
- CSS and Fonts: Inline critical CSS and minimize custom fonts for better performance.
Step 5: “Boldly link where no one has linked before with Text Fragments” – Google
URL Hash Implementation:
Add a URL hash link at the section of the article that directs users and AI systems directly to the quick question and answer section. For instance, if the question “What factors make a website load as quickly as possible?” is summarized as a list because the AI overview generates a list, the URL should be structured to tell the AI systems where the list is on the page.
This ensures that users and AI systems can quickly access the most critical information and is a best practice, according to my research of dozens of AI overview results.
Google could also automatically generate the links based on the search context and the content on our site. This used to only apply to a specific portion of all searches that won the SERP features, but now it seems all URLs will be hashed to highlight the origin text that made up the AI Overview. So, you either hash it, or Google will hash it for you.
Step 6: Re-Index Your Content
When you update a page, submit a re-indexation request through Google Search Console.
Step 7: Off-Page Optimization trumps backlinks
Modern brand reputation management and digital PR strategies will surpass traditional backlink building in AEO world. These strategies focus on boosting visibility and authority through guest posting, answering journalist queries (e.g., Connectively), engaging influencers in your marketing, getting featured in listicles, engaging on platforms like Quora, Reddit, etc., and encouraging positive customer reviews to strengthen brand perception.
What Off Page Optimization techniques are essential to AEO and AI overviews?
- Guest Posting on High-Authority Sites
- Answering journalist queries for potential features and brand recognition.
- Engaging in Influencer Marketing
- Gaining visibility by being featured in authoritative rankings (listicles, awards, etc.)
- Participating in discussions on Quora, Reddit, etc., to build authority.
- Encouraging positive product or service reviews.
Did you see what I just did for step 7?
Navigating the New Frontier of AI Overviews
“So comes snow after fire, and even dragons have their ending!” said Bilbo Baggins after the Battle of Five Armies as he looked towards Erebor with Gandalf. Traditional SEO was a dragon in its might and power over the internet. I don’t know why I’m calling it traditional; it hasn’t left us yet.
The strategies we’ve explored – from understanding the fundamentals of GEO to implementing CSQAF, optimizing content structure, and leveraging off-page techniques – are not just best practices; they’re essential tools for survival and success in this new paradigm.
Remember, the key takeaways are:
- Adapt your content strategy to prioritize direct answers and scannable formats
- Implement structured data and URL hashing for improved visibility
- Focus on authoritativeness, citations, and off-page optimization
- Continuously test, learn, and refine your approach
As content writers, strategists, and SEOs, we must embrace this change while maintaining the human touch that gives our content depth and authenticity. The future of search is not about outsmarting algorithms but about creating valuable, accessible content that serves both human readers and AI systems.
Stay curious, keep experimenting, and don’t be afraid to break away from traditional SEO norms. AI Overviews is still evolving, and those who adapt quickly and thoughtfully will be best positioned to thrive. If you need help optimizing your website for answer engines, our experts can help, feel free to reach out to us.