How to rank in Google AI Overviews and AI Mode: the 2026 guide to earning citations on both surfaces

Here is the short version. Google now runs AI at two scales in search. AI Overviews are the summaries that appear automatically above traditional results, reaching about 1.5 billion monthly users and triggering on roughly a quarter to half of all searches. AI Mode is the separate conversational experience users actively enter for deeper research. They look similar on the surface, but they cite almost entirely different sources, meaning if you optimize for one and ignore the other, you leave half the surface uncovered. And as of May 19, 2026, Google announced they are merging into one seamless experience, so understanding both now gives you a head start on what search is becoming. This guide covers how each surface works, how they differ in sourcing and citation behavior, what earns a citation on each, and a practical playbook for showing up on both. It completes the engine set alongside our guides to ranking in ChatGPT and ranking in Perplexity, and connects back to the full discipline in our answer engine optimization guide.

AI Overviews and AI Mode: two surfaces, one search engine

AI Overviews launched in May 2024 as an automatic feature. When Google determines an AI summary adds value to a query, it generates a concise answer, typically a few sentences to a couple of paragraphs, and inserts it at the top of the traditional results page with citation cards linking to the sources it used. Users do not opt in. The summary appears on its own.

AI Mode is different. It is a separate, dedicated search experience that users actively select by tapping the AI Mode toggle near the search bar. Once inside, the experience becomes conversational: users can ask follow-up questions, explore tangents, and receive longer, more detailed answers with extensive source citations. It uses query fan-out, breaking complex questions into sub-questions and searching across a wider range of sources.

Both are powered by Google's Gemini family of large language models and draw on Google's own search index and entity graph. But the similarities end there. How they select sources, how many they cite, what kinds of content they favor, and how the user interacts with the answer are all meaningfully different.

They cite almost entirely different sources

This is the finding that should reshape your strategy. SE Ranking's study found only 10.7 percent URL overlap and 16 percent domain overlap between AI Overviews and AI Mode when answering the same queries. A separate analysis confirmed that 87 percent of the time, the two surfaces pull from completely different source pools. Even when examining just the top three citations per answer, the overlap increases only marginally to about 16 percent.

The source preferences are distinct. AI Overviews show a strong preference for multimedia content (videos, core pages) and directly answerable, concise content. AI Mode leans toward encyclopedic, detailed textual sources, with Wikipedia appearing in 28.9 percent of AI Mode responses versus 18.1 percent of AI Overviews. Both overwhelmingly prefer article-format content but differ in what they look for next: AI Overviews want a snippet it can lift cleanly, while AI Mode wants depth it can synthesize across sub-questions.

The practical implication is clear: optimizing for one does not transfer automatically to the other. Brands need two distinct but complementary content approaches, concise and directly answerable for AI Overviews, and deep, multi-entity, and comparison-rich for AI Mode, often on the same topic.

Why being cited matters this much

The business case for earning a citation on these surfaces is not abstract. Being cited inside an AI Overview earns 35 percent more organic clicks and 91 percent more paid clicks than not being cited. That is a massive lift on the most trafficked search engine in the world.

At the same time, AI surfaces are absorbing clicks. Studies show CTR drops of 15 to 70 percent on queries that trigger an AI Overview, depending on the study and query type. AI Mode specifically shows a roughly 93 percent zero-click rate, where users get the answer, scan the citation chips, and leave without clicking through. That sounds dire until you flip the framing: if your brand is cited consistently in AI Mode for the queries your buyers run, the citation itself shapes their perception and consideration set, even when they do not click. And when they do click, the traffic converts well, with Similarweb finding AI referral traffic converts at about 7.1 percent.

The strategy is not to fight the zero-click trend. It is to earn the citation.

What earns a citation in AI Overviews

AI Overviews are built for speed and concision. They trigger most on informational, how-to, and best-of queries where a clear consensus exists across multiple authoritative sources. The content they cite shares a consistent profile.

Featured-snippet-style clarity

AI Overviews evolved directly from featured snippets, and the optimization discipline is similar. They reward content that delivers a direct, complete answer in the first 40 to 60 words of a section, formatted as a clean, liftable passage. If you have been optimizing for featured snippets, much of that work carries over.

Concise, directly answerable structure

Because the summary is short (typically a few sentences to a couple of paragraphs), the engine favors pages where the answer is immediately available, not buried in a long-form guide. Clear headings that mirror the query, short paragraphs, and standalone passages that work if extracted in isolation all increase citation probability.

Traditional SEO authority

AI Overviews sit on top of Google's index, so your traditional SEO foundation, domain authority, backlink profile, E-E-A-T signals, and crawlability, carries over more directly here than on any other AI surface. Pages that rank well in traditional Google results have a meaningful advantage in AI Overviews, though ranking is not a guarantee of citation.

Structured data and entities

Both surfaces rely on Google's entity graph, but AI Overviews lean particularly on structured data to understand what a page is about. Schema markup (Article, FAQPage, Product, HowTo, Organization) helps the AI parse entities and match content to queries more confidently. E-E-A-T is described as non-negotiable by multiple analyses: if your site lacks clear signals of experience, expertise, authoritativeness, and trustworthiness, the AI is unlikely to cite you.

Multimedia

AI Overviews show a distinctive preference for multimedia content, including video. Pages with embedded, relevant video or visual content are cited at higher rates than text-only pages for applicable query types. This is a differentiator from AI Mode, which favors pure textual depth.

What earns a citation in AI Mode

AI Mode is built for depth and exploration. Users actively enter it for more complex, multi-step queries, and the engine responds with longer, more detailed answers that reference more sources and more entities. The content it cites is different.

Deep, multi-entity content

AI Mode references about 3.3 entities per response versus 1.3 for AI Overviews. It favors content that covers multiple related entities, such as comparison pages, multi-product roundups, and deep guides that connect concepts. A page that only answers one narrow question may be fine for AI Overviews but too thin for AI Mode's synthesis.

Encyclopedic and detailed textual sources

Wikipedia appearing in 28.9 percent of AI Mode responses signals what the surface values: comprehensive, well-organized, factual reference material. Your content does not need to be Wikipedia, but it should share the traits that make Wikipedia citable: clear structure, thorough coverage, neutral tone, and sourced claims.

Passage-level relevance

AI Mode evaluates content at the passage level, not the page level. It uses query fan-out, breaking complex questions into sub-questions and retrieving passages from different pages for each. This means a single deeply relevant passage buried in a larger guide can be cited, but only if it is well-structured enough for the engine to find and extract it. Every section should be independently answer-first, even inside a comprehensive guide.

Comparison and decision-support content

Because AI Mode handles more complex queries, comparison content, decision frameworks, and versus pages earn disproportionate citations. Content structured around when-to-choose-X-versus-Y, with clear comparison tables and specific use-case guidance, aligns directly with the kind of queries that drive users to AI Mode.

Freshness

Freshness matters on both surfaces, but AI Mode's deeper synthesis means it is more likely to pull in recently updated content to support a multi-faceted answer. The freshness principles from the broader AEO research apply: pages updated within 60 days are 1.9 times more likely to appear, and content refreshed within 30 days earns roughly 3.2 times more citations.

The May 2026 merger and what it means

On May 19, 2026, Google announced that AI Overviews and AI Mode are being merged into one seamless AI Search experience. The details of how the merged surface will work are still evolving, but the direction is clear: the distinction between a static summary and a conversational experience is collapsing into something more fluid, where a quick answer can expand into a deeper conversation within the same interface.

What this means for your strategy is that the optimization work you do for both surfaces now is not wasted, it is preparation for the merged experience. Content that is concise enough to be lifted as a quick answer and deep enough to support a follow-up conversation is the profile that wins in both the current split and the merged future. The brands that can serve both modes from a single, well-structured content library will have a structural advantage.

A practical playbook for both surfaces

1. Audit your current visibility on both. Run your top 30 to 50 buyer queries through standard Google search (to check AI Overviews) and through AI Mode separately. Record whether you appear on each, which surfaces cite you, and where competitors show up instead. The 87 percent source divergence means you may be strong on one and invisible on the other.
2. Build two-tier content for key queries. For your highest-value topics, structure content so it works at both scales. Lead each section with a direct, concise answer that AI Overviews can lift (the first 40 to 60 words), then expand with the depth, entities, comparisons, and sourced data that AI Mode needs. Think of each section as a featured snippet on top and a pillar section underneath.
3. Strengthen traditional SEO. Because both surfaces draw on Google's index, your traditional authority, backlinks, E-E-A-T, crawlability, and structured data carry over more directly here than on ChatGPT or Perplexity. Shore up the foundation.
4. Add comparison and multi-entity content. For AI Mode specifically, build comparison pages, versus content, and guides that connect multiple entities. These earn citations disproportionately on the deeper surface.
5. Use schema deliberately. Implement FAQPage, HowTo, Article with author, Product, and Organization schema. Both surfaces use Google's entity graph, and structured data helps them parse and trust your content.
6. Consider multimedia for AI Overviews. If applicable to your category, add relevant video or visual content to key pages. AI Overviews cite multimedia at higher rates than AI Mode does.
7. Refresh on a cadence. Both surfaces reward freshness, but with 40 to 60 percent of AI Overview citations rotating month over month, a stale page is a constant risk. Refresh high-value commercial pages at least quarterly.
8. Monitor both surfaces separately. Because they cite different sources, tracking AI Overviews alone misses your AI Mode visibility and vice versa. Track share of voice on each, and watch for the merger's rollout to see how citation behavior shifts.

outwrite.ai connects research, writing and scheduled publishing with visibility measurement. Results samples ChatGPT, Gemini and Google AI Mode and keeps mentions, recommendations and citations separate. Optional Google Search Console data adds your own search performance to planning. Use both kinds of evidence to guide the work; an AI mention is not the same as an organic visit or a sale.

Where outwrite.ai fits

Check each search surface with the appropriate method. outwrite.ai currently samples ChatGPT, Gemini and Google AI Mode, normally roughly weekly. It does not currently provide Google AI Overviews monitoring in Results. Use the Results guide to understand the supported coverage, and investigate AI Overviews separately when that is the channel you want to assess.

The bottom line

Google AI Overviews and AI Mode are the highest-reach AI search surfaces in the world, and they behave like two almost entirely separate systems when it comes to choosing sources. Optimizing for one and assuming it covers the other leaves roughly half the surface uncovered. AI Overviews reward concise, directly answerable, snippet-style content backed by traditional SEO authority. AI Mode rewards deep, multi-entity, comparison-rich content evaluated at the passage level. Both reward freshness, structured data, E-E-A-T, and entity clarity. And with the announced merger, the brands building content that works at both scales now will be structurally ready for what Google search is becoming. If you want to see where you stand on both surfaces, track how your citations shift over time, and start earning the spots your competitors are taking, that is what outwrite.ai was built for.