How to Get Featured in AI Overviews (Based on Data)

AI Overviews have thrown marketers into a tough spot. On one side, Sundar Pichai, Google’s Chief Executive Officer, insists that AIOs send traffic to a wider set of websites. However, there is palpable fear among marketers and publishers, with reports of declining click-through rates (CTR). 

After analyzing conflicting studies and talking to search engine optimization (SEO) professionals, I believe the truth, as it often is, is far more nuanced. This isn’t a traffic apocalypse as some fear. It’s a restructuring of search. As a Siege Media report puts it: “The sky isn’t falling. It’s shifting.”

The deterministic world of SEO, which is dependent upon keyword optimization and acquiring links, is making way for a probabilistic one. The new reality is governed by an opaque, ever-shifting logic behind generative AI.  

In the following sections, we will cut through the noise. First, we’ll understand the new search engine results page (SERP) economy to understand where our old metrics are failing us. Then, we’ll dissect AI citations to study who gets chosen and why. Finally, we’ll discuss actionable shifts your team must make to thrive in this new era.

Decoding the new SERP economy

Based on my analysis, the old search economy is upended. Traditional metrics are unable to capture an emerging search economy, where attention, not just rank, is a scarce resource. Here are the changes that define this new reality:

1. Why ranking alone is not enough

The number one spot on SERP is no longer a reliable proxy for visibility. The SERP’s center of gravity is shifting towards AIOs at the top from the blue links. The new critical metric for Google is attention, which is exemplified by visual placement. According to a joint report by Botify and DemandSphere, AIOs can take up to 48% of the screen space on mobile phones. Organic results are now moving to be functional footnotes, catching attention only when the top view, containing AIOs, doesn’t satisfy intent.

Visibility now depends on:

1. Placement inside the AI summary
2. Content authority signals
3. Semantic alignment with user intent

“The top trust signal for buyers across the board was explainability — how the technology works.”

A Leap of Trust: AI Agents Are Winning Hearts and Wallets

2. Movement from discovery to validation 

The purpose of clicks is more diversified now. AIOs provide an initial summary, so the user’s subsequent click is no longer an act of open-ended discovery. My analysis points to it being an act of validation. This is more evident in high-consideration stages along the buyer journey. A TrustRadius study suggests that 90% of B2B buyers click through on AIO citations, with the probable intention of fact-checking. This also reveals a trust deficit in generative AI responses. 

AI sources can introduce information, yet users are hesitant to trust them completely. It takes the authority of first-hand human-generated information. Landing pages today shouldn’t just answer initial questions. They must provide deeper context than the AI summary. Your content must be built to earn trust.

As the G2 AI Agents report notes, “A trust gap grows when reliability improves faster than adoption.” This indicates that users require confirmation before accepting automated outputs. According to the report, software buyers are “more resolute than ever in vetting agentic solutions for performance and risk,” actively cross-checking claims before accepting them. 

This means your content’s job is to:

1. Provide depth beyond the AI summary
2. Establish credibility and firsthand expertise
3. Answer why something matters, not just what it is

3. Clicks as a two-tiered system

AIOs are not uniformly killing clicks. In fact, they are creating a divide between two classes of pages: cited and unseen. The outcomes are visibly harmful for the unseen pages. A study by Terakeet, as reported by Search Engine Land, has revealed that being excluded has “measurable and significant damages for the webpage.”

A valuable opportunity has emerged for the cited. The same study found that webpages included in AIOs have 3.2x as many clicks as the excluded pages.

4. Influence as a new core KPI

The old SEO model was transactional. The new one is about influence. That’s the only way to explain why overall clicks are down, but clicks to homepages (as the Siege Media report above shows) are up 10.7%. We are witnessing a new user journey. 

The AIO acts as a digital concierge. It does the initial research and pulls out a few trusted brands. The user then either makes a validation click or directly heads to a specific source using brand recall. SEO, thus, must now be viewed as a brand marketing channel. Success must be measured more in terms of branded search volume and direct traffic, which are the true signals of your brand having positively impacted the top of the funnel. 

The new SERP economy is about inverted incentives and abstract value. A single qualified click can be worth more than a dozen clicks. To make the most of it, marketers need a clear, actionable path forward. The playbook below offers them a way to stop reactively engaging with evolutions in SERP and start proactively shaping their presence within it.

How to show up in Google's AI Overviews: a playbook

The biggest mistake a marketer can make now is to consider AIOs as another feature to optimize for. It’s not an update but a profound restructuring of how information is delivered on the internet. Adapting to this shift requires a new strategic framework. Although there are no defined rules guaranteeing a position in AIO, certain patterns emerge after analyzing the results. You, too, can tap into AIOs’ potential by adopting my top five research-backed practices: 

1. The nuanced authority mandate

The first principle is about authority, but it’s not defined by rank alone. As Senior SEO specialist and G2 Icon Kishan Soni says from his experience, “AIOs and generative AI search are prioritizing conversational and authoritative context-rich responses...generative AI tools pull from high-quality, trustworthy sources to deliver direct answers”. 

The data on what constitutes a “trustworthy source” is contradictory, which means the algorithm weighs multiple factors for this. A study by seoClarity found that one or more of the top 10 results are included in AIOs 99.5% of the time. Whereas another study suggests that nearly half the citations are pulled from beyond the first page. 

This shows that while ranking is still the most powerful signal, it’s not the only one. My interpretation suggests that Google uses high rankings as a baseline for trust, but for certain queries, it’ll defer to results beyond page 1. Domain authority is gaining more prominence now. 

2. Content-as-answer blueprint

The algorithm operates on frequency bias. It favors information that is widely repeated, cited, or accepted as credible across the internet. You must structure for machines and write for humans. Based on my research, this means: 

A. Lead with answers: Address the core question immediately. Don’t distract the reader or AI. To adapt to AIOs, SEO expert and G2 Icon Deyan Georgiev started treating every page like a potential answer source for AI, not just people. “That means opening content with direct, natural-language answers to likely queries, almost like I’m writing for voice search,” he tells me. Another G2 Icon, Mayra Shaikh, who’s also an event tech specialist, emphasizes she uses “answer-first content”.

B. Structure for scannability: Use structural elements like question-based headings, bullets, and concise paragraphs, not just for a human reader but also for an AI parser. Mayra says she uses structured data and schema markup and optimizes for long-tail, conversational queries.

C. Product blogs: My analysis of several responses from AIOs found that B2B-related queries provide a valuable opportunity for vendor-owned content to be cited directly. This is because for many niche B2B topics, authoritative third-party content is rare. 

By creating high-quality blogs, companies can fill this gap and become go-to sources for AI engines. A Rankscale.ai study has found vendor-authored product blogs saw a notable inclusion of 7% in AIOs. These include blogs on product features, comparisons, and guides.

"We've shifted from purely 'how-to' content to more inspirational guides."

Prabhat Singh
Lead of Site Strategy and Planning at a B2C firm

3. Focus more on UGC

Platforms like Reddit and Quora have seen explosive growth along with AIOs and are among the dominant sources there. An Ahrefs study has found that websites showing up the most in AIOs include Wikipedia, Reddit, and Quora. 

This represents the algorithm’s attempt to find and reward “experience” in E-E-A-T. Andy Crestodina, Chief Marketing Officer at Orbit Media Studios, notes Reddit was "brilliant" at training data for AI because it contains a wealth of "conversation and opinion, and vernacular common language. " He believes Google now prioritizes it because it’s “very likely to be satisfying to the visitor”. The implication is clear: You must engage users where conversations are happening. Your presence and credibility across these platforms are now signals of authority. 

To build “experience authority”:

1. Participate in relevant subreddits
2. Summarize case examples in community spaces
3. Repurpose audience Q&A content into structured web articles

According to G2’s A Leap of Trust: AI Agents Are Winning Hearts and Wallets Report, “Large enterprise buyers rated vendor-provided internal reliability metrics as the least trustworthy signals. To win them over, you’ll need to bring third-party proof.” Even AI increasingly draws from third-party sources, reviews, and community knowledge, not just from top-ranking pages.

4. Reverse engineer LLM logic

Ross Briggs, Vice President of Buyer Experience at G2, points out how an analysis of retrieval augmented generation (RAG) for AI platforms reveals Bing is emerging as a main source for ChatGPT, Claude, and Perplexity.

“For large language model (LLM) inclusion, users must reverse engineer results to examine the logic used and sources tapped”

Ross Briggs
Vice President of Buyer Experience at G2

For this, first enter your target prompt in a platform and examine the page type it pulls for results. Next, use your metadata to appeal to LLMs. Then, focus on cosine similarity, which means aligning your content to the question asked in the LLM for semantic relevance. 

Another part of this is audience prompt engineering, which is about tracing conversational journeys of your ideal customer profile (ICP). Ross says G2 is approaching this by combining ICP personas with long-tail keyword data: “Then we use AI to form different likely prompts that buyers are searching for." 

This shifts the emphasis from what the audience is searching for to what they would search for. This needs a deeper understanding of your audience’s queries, pain points, and branches of sub-queries to satisfy the “fan-out” method used by AIOs.

5. Treat every page as a front door

With AIOs, it’s not just product pages and the homepage that matter the most. AIOs act as a universal key, capable of unlocking your brand through different entrances. Every article, product guide, or FAQ matters more than before. BrightEdge has found that 82.5% of citations are linked to deep content pages, which are two or more clicks away from the homepage.

AIO’s intention is to provide useful information to the user, regardless of your site’s navigational architecture. It can bypass user funnels you’ve created. Every piece of content must carry its own weight, establish its credibility, and be ready to serve as a user’s first, and maybe the only, interaction with your brand. 

Remember to consider turning your valuable gated content, such as case studies and whitepapers, into blogs. TrustRadius reports that AI models can only read publicly accessible data for training purposes. “Vendors must find the right balance between gated and ungated content to maintain discoverability in the age of AI,” the report recommends. 

Discovery on AI platforms is the new competitive frontier for brands. Watch Reach 25, where the industry’s brightest minds shared how to show up first, earn trust fast, and turn every AI moment into unstoppable brand momentum.

Checklist: How to show up more in AI Overviews

Start optimizing for influence 

AIOs and generative AI search are not just reshaping search marketing. They’re altering power dynamics among brands, platforms, and consumers. The coming years will see the emergence of AI-native marketing teams that build their entire content and SEO strategies around machine readability from Day 1. 

The value of basic informational content is plummeting as AIO marches ahead. To survive this transition, brands must master contextual storytelling. Andy makes the distinction between algorithm marketing and relationship marketing, which includes building PR, harnessing influencers, and enabling word-of-mouth. He argues that in the age of AI, relationship marketing becomes paramount as it’s how your brand is expressed in the world. 

Marketers shouldn’t stop at answering a series of disparate queries for SEO. They must conduct semantic content mapping to understand not just what the audience asks, but also the web of surrounding queries. The winners will be those who feed AI systems with anticipated answers to questions users haven’t asked yet. 

Here's the immediate action you can take: 

• Audit your content using an AI lens. This means going beyond keywords and examining content for logical flow, credibility, and factual density. More importantly, it means reviewing the cosine similarity of content. 
• Set up a direct relationship with your audience through owned channels, as AI-mediated discovery reduces dependence on your platforms. 
• Invest in content optimization tools that can adapt your messaging as the AI algorithm evolves. These tools must offer a unified view of your content’s discovery across channels, including AI platforms. 

The emergence of AIOs creates a dilemma for us — should we create content for humans or machines? True brands that stand out and thrive in this era will offer unique views, balance traditional and new SEO practices, and dominate communities so that AI systems have no choice but to reference them. Not because they’ve gamed algorithms, but because they’ve become indispensable to the conversation. 

AI-backed cyberattacks are more targeted and lethal. How are defenders responding? They are harnessing AI to tackle AI-powered threats. Learn more.

FAQs

1. Does ranking higher in SERP guarantee citations in AI Overviews?

Not necessarily. While top-ranking positions act as an important trust baseline, AI Overviews also evaluate factors like domain authority, relevance, clarity of explanations, and presence in user-generated discussions. For many queries, citations are pulled from beyond page one when those sources demonstrate stronger contextual depth or practical experience.

2. How do I write content that AI Overviews are more likely to cite?

Write for both human readability and machine interpretation. Open pages with direct, natural-language answers, then expand into structured explanations with headings, bullets, and clear reasoning. Use question-based section titles and include examples, comparisons, and definitions that map to common queries. Also, ensure your content is consistent across platforms. AI systems cross-reference your claims against your digital footprint.

1. Should I shift from a keyword-centric strategy because of AI Overviews?

Keyword strategy should shift from exact-match targeting to semantic intent targeting. AI systems evaluate content by meaning, not only phrasing. This requires focusing on a series of related questions, variations of user intent, and contextual explanations. Instead of optimizing for one keyword, optimize for the conversation surrounding keywords.

Buyers are turning to generative AI for software search and selection. Learn how to build a brand that AI can’t ignore. Watch Reach 25. 

Edited by Supanna Das