The Future of SEO 2025: Strategy Optimization (AEO) for the Google AI Era Overviews

PART I: AI OVERVIEWS AND THE ZERO-CLICK CRISIS: IMPACT DATA ANALYSIS

I.1. Background: The Rise of Generative Search Experiences (SGE/AI Overviews)

The integration of Generative AI into Google Search, known as AI Overviews (AIO) or Search Generative Experience (SGE), marks a fundamental shift in the way users interact with online information. This model is more than just an added feature; it is a platform transformation from a traditional search engine (Search Engine) to an answer engine (Answer Engine). This radically changes the rules of visibility and traffic distribution.

The frequency of AI Overview appearances is increasing at an alarming rate, showing that Google is accelerating the integration of AI into the core search experience. Data shows that AIO is present in 30% of desktop queries in the US as of September 2025—a record high. More notably, the frequency of AIO appearances on mobile devices has increased by 474.9% year-on-year. This growth forces SEO strategists to act quickly. SGE operates on a real-time adaptive mechanism, similar to a "chef adjusting the menu based on seasonality and the freshest ingredients." This means search results can change based on the time of day, trending topics, or specific user behavior. This volatility increases the complexity of maintaining traditional rankings.

I.2. Analyzing the Impact of Zero-Click Search and Reducing CTR

The biggest impact of AI Overview is the dramatic increase in the Zero-Click Search phenomenon, where users get the answers they need right on the search results page (SERP) without needing to visit any external websites. These answers often come from featured snippets, knowledge panels, or AI Overviews.

Statistical data supports this concern. Currently, about 58.5% to nearly 60% of all searches performed on Google, both on desktop and mobile platforms, end without any clicks. Some analysts even predict that this share could surpass the 70% mark by the end of 2025 as AI summaries and generative features become more popular.

The direct impact on traffic is clear. Ahrefs' independent research on 300,000 keywords shows that the presence of AI Overview is directly correlated with an average 34.5% reduction in click-through rate (CTR) for traditional results. This massive decline coincided with large-scale AIO deployments, leading experts to call AIO a "crocodile mouth" that is eating away at traditional organic traffic.

This challenge is especially acute for small and medium-sized businesses (SMBs) as well as top-ranking links, who are facing unprecedented difficulty in driving traffic to their websites.

To better visualize this shift, the following summarizes the key quantitative impacts:

Quantitative Impact Analysis of AI Overviews and Zero-Click Search

Zero-Click Search Rate:
- Current Statistics (Estimated 2024-2025): Approx. 58.5% to 60%.
- Strategic Implications: The level of dependence on direct SERP increases, requiring optimization for display (Visibility) instead of Click (Traffic).
CTR reduction with AIO:
- Current Statistics (Estimated 2024-2025): Average reduction 34.5%.
- Strategic Implications: Direct challenge to the traditional traffic-based business model.
Frequency of AIO (Mobile):
- Current Statistics (Estimated 2024-2025): Increase 474.9% YoY.
- Strategic Implications: The need to have a mobile optimization strategy and AIO as a top priority.
AIO Citation Source:
- Current Statistics (Estimated 2024-2025): > 99% from Top 10 web results.
- Strategic Implications: Even though Clicks are down, maintaining a Top 10 ranking is still a prerequisite to get cited by AI.

I.3. Insight: The Difference Between Visibility and Traffic

The data on the decline in CTR and Zero-Clicks should not be interpreted as SEO going out of style, but as a signal of a fundamental shift in the goals of SEO: from Traffic SEO to Authority SEO.

The analysis reveals an important paradox: even though users do not click after See AIO, being displayed in AIO depends almost entirely on traditional rankings. Studies show that more than 99% of AI Overview versions are compiled and quoted information from Google's top 10 results.

This leads to a strategic conclusion: to be cited by AI (Visibility), the website must still satisfy Google's core ranking algorithms. If a business is not in the Top 10, they will not exist at all in the AI's aggregated answers. Therefore, the goal of SEO is no longer to redirect traffic (Traffic Router) but to verify and provide authority (Authority Verifier) to the Large Language Model (LLM).

This difference between visibility (Visibility) and traffic (Traffic) is key. Being chosen by Google as a citation source still brings great value to the brand. Appearing prominently in Featured Snippets or AI Overviews helps build brand awareness, strengthen credibility, and establish that brand as a trusted source in the industry.

PART II: DECODING THE FIVE DIMENSIONS OF NEW SEARCH IN THE AI ERA

SGE is reshaping user behavior, going beyond the single keyword query model. These five new search styles require specialized optimization strategies.

II.1. Conversational Search

Powered by Generative AI, today's search experience is more conversational and supportive than ever. Users no longer enter static keywords but instead use long-tail, question-based keywords and conversational queries, allowing them to refine, follow, and discover information naturally.

This change requires a mindset shift for strategists: instead of searching for "bike repair tips," users ask "How can I fix my bike chain?". To be responsive, content needs to be written in a natural, spoken-language style, and provide direct answers to specific questions in the FAQ section or at the top of the page. Tools like AnswerThePublic or AlsoAsked become important for discovering the actual questions your target audience is asking.

II.2. Multimodal Search

Multimodal search involves combining text, images, voice, and chat in a more seamless and intuitive search journey. Google's MUM (Multitask Unified Model) is connecting these types of data, making visibility dependent not only on text but also on how image and video elements are indexed.

To optimize for this dimension, businesses like Tan Phat Digital need to ensure that all visual content is backed by strong technical data. This includes adding detailed and accurate Alt Text descriptions to images, embedding videos with clear transcripts, and optimizing videos to answer specific questions. Additionally, maintaining logo and brand image consistency across different platforms is necessary to strengthen the brand entity.

II.3. Action-Oriented Search

Action-Oriented Search focuses on completing a specific task or job (buyer tasks). Users look for instructions, steps, or products/services to achieve an end goal.

AI prioritizes clearly structured content that can easily be extracted into sequential steps. Content developers need to use specialized Schema Markup like HowTo to encode instructional steps, allowing AI to easily extract and display the process. This not only helps Rich Snippets but also "grounds" the action logic for the LLM, increasing the likelihood of being cited in complex tutorial sequences. Additionally, providing verifiable evidence (e.g., screenshots, worked-out examples) and encoding meaning with structured data like FAQ or SoftwareApplication is important so that the system can trust and cite the source.

II.4. Synthesis Search

AI Overviews acts as a tool to synthesize, compare, and distill information from multiple authoritative sources instead of just answering with a paragraph from a single source. This shift requires enterprise content to be built to become independent blocks of information that can be "lift-and-placeable" into AI's aggregated answers.

To be successful in comprehensive search, the page structure must achieve "modular clarity". This is done by: using clear headings (H2 for main ideas, H3 for details) to divide each idea; Use lists, tables, and bold formatting to highlight core facts. Well-divided content not only assists readers but also helps AI identify related passages quickly.

II.5. Personalization Search

In the AI era, users don't just want simple answers; they want a "trusted advisor" who can anticipate needs and remember search context. The AI tries to adapt to the user based on time, location, and previous behavior.

Relevance and personalized context are increasingly more important than keywords alone. This enhances the role of Local SEO, because nearly half of voice searches are local searches. The adaptive strategy includes optimizing your Google Business Profile, ensuring NAP (Name, Address, Phone) consistency across the entire online network. More importantly, building semantic content clusters (Semantic Content Clusters) helps AI clearly understand the scope of expertise and topical authority (Topical Authority) of a brand in a specific field.

PART III: STRATEGIC PILLAR 1: CONVERTING CONTENT AND AEO

The first and most important pillar in an AI-era SEO strategy is the transformation of the content production model, focusing on Answer Engine Optimization (AEO) instead of traditional SEO.

III.1. From Keyword Targeting to Answer Targeting

Traditional SEO focuses on keywords and phrases to attract clicks. In contrast, LLM Optimization (or AEO) focuses on providing direct and complete answers to be cited by the AI in its aggregated responses.

This difference in goals leads to differences in structural requirements:

Start with the Answer: Content should begin with a direct, succinct response (ideally 40-60 words) right below the main heading (H1 or H2) to immediately satisfy user intent. This maximizes your potential to become featured snippet or cited in AIO.
Logic Structure: Use clear headings and subheadings (H2, H3) to divide content. Each section should keep a single idea or topic, avoid mixing multiple topics in the same paragraph. This structure makes it easy for AI search engines to read user intent, map entities, and locate the correct answer quickly.

Comparison of Optimization Framework: Traditional SEO and LLM Optimization (AEO)

Main Objective:
- Traditional SEO: Focuses on Ranking Links and attracting clicks.
- LLM Optimization (AEO): Focus on being Cited in AI answers.
Optimum Unit:
- Traditional SEO: Focuses on single Keywords and keyword phrases (Keywords).
- LLM Optimization (AEO): Focuses on Entities, Answers, and Topical Depth.
Trust Signal:
- Traditional SEO: Relies on Backlinks and Domain Authority.
- LLM Optimization (AEO): Based on Evidence of Expertise, Proprietary Data, and Authenticity (E-E-A-T).
Priority Structure:
- Traditional SEO: Long Segmentation, Keyword Density.
- LLM Optimization (AEO): Direct answer 40-60 words, lists, tables, schema markup.

III.2. Conceptual Framework Query Fanout and Topical Depth

In generative search environments, the Large Language Model (LLM) goes beyond the user's initial query. They perform a process called Query Fan-Out (QFO), where a single query is analyzed and broken down into multiple sub-queries to explore a broader or deeper set of answers. For example, a search for "best protein for runners" could be expanded into queries like "best whey protein," "plant-based protein," or "post-run recovery supplements."

The QFO concept places an absolute requirement on Topical Depth. LLMs don't just look at individual keywords; they evaluate whether the site covers the entire topic cluster. If a website only has Pillar Content but lacks supporting articles for secondary queries, AI will conclude that the website lacks expertise and depth. This forces strategists to move from single keyword hunting to building comprehensive content systems.

To effectively implement QFO, take the following steps:

Map Content Clusters: Identify Pillar Articles as the main comprehensive guide, then create Subtopic Pages (e.g., comparisons, use cases) and Supporting Content (FAQs, case studies).
Build Semantic Completeness: Content should include all relevant subqueries, not just the surface level.
Tightly Linked: Use internal linking to connect assets within the cluster, helping the AI understand hierarchy and depth of expertise.

III.3. Protect Content with Proprietary Data

The biggest threat to traditional SEO is AI Content Cannibalization—a condition where AI aggregates, copies, and rewrites generic content into competing versions, diluting the original SEO signal and leading to decreased rankings.

To protect content and strengthen the E-E-A-T signal, businesses like Tan Phat Digital must focus on creating assets that AI cannot replicate or synthesize. AI works by synthesizing existing data; it cannot create or summarize data that does not yet exist.

The most effective defense strategy is to focus on publishing Proprietary Data:

Original Research and Surveys: Conduct internal research, run surveys, or analyze proprietary data to provide fresh insights to the industry. For example, Tan Phat Digital can publish "Paid Media Advertising Performance Index Report in the Vietnamese market".
Case Studies and Real Results: Share specific results and figures that reflect the actual work or observations of the business.
Interactive Data and Tools: Publish original data tables, charts, ROI calculators, or tools that AI can only cite but not create.

III.4. Real-Life Example: The Advantages of Proprietary Data (Case Study)

Proprietary data-driven content is not only an SEO tactic but also an incredibly effective branding and link building strategy. The advantage of this type of content is low copyability, freshness and ease of attracting press attention.

Below are illustrative examples of the types of Proprietary Data that Tan Phat Digital can apply:

Industry Survey Report: Similar to how Thumbtack conducts its Annual Small Business Friendly Survey. Tan Phat Digital can conduct "Survey of Digital Marketing Spending Trends of SMEs in Vietnam" and publish indicators by province or industry.
Proprietary Calculation Tools: Marketing companies like WebFX have built proprietary platforms and tools (e.g., ROI calculators, ad quality score calculators) to help clients calculate performance. This is a high-value asset that is cited by AI and attracts potential customers.
Analyze Original Internal Data: Comparative studies like Hubspot's "Plain Text Email Performance vs. HTML Email" provide exclusive insight. Tan Phat Digital can create similar analytics on the performance of real campaigns (e.g. "Best Engagement Rate for Facebook Carousel Ad Format in Vietnam").

Proprietary data provides “Experience” and “Expertise” signals that cannot be copied, builds high barriers to competition, and forces AI to cite the original source when summarizing new findings.

PART IV: STRATEGIC PILLAR 2: STRENGTHENING TRUST AND AUTHORITY (P.E.E.A.T)

In the answer engine era, trust is the most important ranking factor. Google has integrated E-E-A-T (Experience, Expertise, Authoritativeness, Trustworthiness) signals into its core ranking system and Knowledge Graph to determine which sources are most trustworthy for AI Overviews.

IV.1. E-E-A-T: The Determinant for Citation

E-E-A-T is more than just a simple ranking factor; it is a prerequisite for recognition by AI. The E-E-A-T signal directly affects which website AI Overview chooses to synthesize information on. Brands that demonstrate strong E-E-A-T have a better chance of attracting higher quality and more confident visitors, because they trust the results compiled by AI.

The transition to SGE and recent core algorithm updates show a clear preference for quality and user-centricity, over traditional SEO tactics. Google uses manual quality ratings (Search Quality Raters), adversarial testing, and fact-checking systems to ensure AI Overview prioritizes trustworthy information. Brands lacking proven credentials (especially the Knowledge Graph and other reputation metrics) will likely have serious difficulty appearing in AI-powered search experiences.

IV.2. Clearly Displaying Experience and Expertise

To satisfy the E-E-A-T criteria, displaying evidence of Experience and Expertise must be a priority:

Experience: Must be demonstrated through evidence of time in the industry and actual involvement in the subject topic. For Tan Phat Digital, this means sharing real-world results, worked examples, and data based on real-world experience.
Expertise: Need to present content with a unique perspective, avoid content easily created by AI. Presenting in-depth and unique perspectives from a team of experts is a way for Google to recognize content as unique and valuable.
Authorship: Enhancing author information is an important technical strategy. Make sure all important content has the author, reviewer, and update date clearly visible. This helps AI confirm authenticity and enhance E-E-A-T.

IV.3. Harnessing Trust Signals through Citations and Freshness

Trust Signals are the foundation on which LLMs build their answers. To increase Trust:

Citations and Evidence: Support factual claims with verified sources, especially data from outside organizations or proprietary studies. Placing these sources near factual statements helps the AI confirm E-E-A-T.
Freshness: Google and AI search engines reward regularly updated content. Keeping content fresh and current with insights gives it priority over static AI copy. This includes updating Schema Markup when content changes so the AI doesn't cite outdated information.

PART V: STRATEGIC PILLAR 3: TECHNICAL ARCHITECTURE AND ENTITY SEO

Entity SEO and Schema Markup are the technical underlayments to effectively communicate with LLMs, establish a clear brand identity, and ensure AI can trust the source of the brand's information.

V.1. Brand Entity Defining

Entity SEO is the optimization process by which search engines recognize a brand—such as Tan Phat Digital—as a distinct entity and associate it with related concepts and topics in the Knowledge Graph. Unlike traditional SEO that focuses on keywords, Entity SEO focuses on clarity, authority, and context.

The main goal of Entity SEO is to have a Knowledge Panel—a Google-verified information box that provides structured information about the entity. Achieving the Knowledge Panel is an official confirmation for Google and LLMs.

The Correlation between Entity Inconsistency and AI Invisibility: LLMs rely on Knowledge Graph to verify the authenticity of entities. Knowledge Graph needs consistent data from multiple external sources (NAP, Schema, Citations) to confirm the legitimacy of the Entity. If Entity data is fragmented or out of sync (for example, NAP information does not match), the AI cannot confirm the authenticity of the Entity, resulting in the brand being considered an unclear Entity and resulting in "AI Invisibility"—not cited at all in the AI's response.

To strengthen Entity Core for Tan Phat Digital:

NAP Data Consistency: Ensure Name, Address, and Phone Number (NAP) is consistent across all digital platforms: website, Google Business Profile, industry directories, and social media channels.
Strengthen External Citations: Build a profile on directories, review platforms, and authoritative external sources to reinforce the Entity's legitimacy.
Identity Synchronization: Ensure logos, images, and brand names are consistent across all channels, including organizational schema, to create a unified identity.

V.2. Optimizing Schema Markup for AI Overviews

Schema Markup is structured data added to a website to help AI Overviews easily identify, extract, and cite your content. AI doesn't want to have to speculate about content; it needs clean and clear signals to understand what the answer is, who the author is, and when the content was published.

Using Schema significantly increases the likelihood of being cited in AI Overviews, thereby increasing visibility and credibility even when users do not click.

Critical Schema Types for AIO:

FAQ Schema: Makes it easy for AI to pull core questions and answers into generative results.
HowTo Schema: Provides a clear step-by-step structure, especially important for Action queries.
Organization/Author Schema: Provides context about the author/reviewer and organization, reinforcing E-E-A-T and Trust.

Implement the Schema Technique:

Use the JSON-LD format, as Google recommends it as easier to implement and less error-prone.
Only mark up content that is visible to the user. Hidden or misleading markup may result in penalties.
Check and update Schema regularly using Google's Rich Results Test to ensure accuracy and synchronization with current content.

V.3. Instructions for Building Knowledge Panel for Tan Phat Digital

Setting up Knowledge Panel is a step to confirm the Entity for Tan Phat Digital in Google's Knowledge Graph.

Specific Action Plan:

Create and Verify a Google Business Profile (GBP): Tan Phat Digital needs to ensure the GBP profile is created, verified, and optimized with accurate contact information, hours of operation, and website links. GBP is a powerful free tool for managing local brand information.
Sync Images and Logos: Choose high-resolution images and a clear logo that is consistent with the brand. Make sure the logo is optimized (square, on a plain background) to display crisply on Knowledge Panel and devices.
Boost Offsite Reputation: Encourage happy customers to leave positive reviews on Google Reviews and other review sites.
Implement Organization Schema: Implement Organization Schema consistently across the entire website, providing branding details and links to official social media profiles.
Monitor and Update: Regularly check and update information on Knowledge Panel, as Google tends to update active profiles.

PART VI: STRATEGIC PILLAR 4: CONVERSION OPTIMIZATION (CTA)

VI.1. Converting Visibility into Business Value

The new goal of SEO is not to reduce Zero-Clicks, but to leverage them. The Zero-Click Crisis forces marketers to change the way they measure the value of visibility.

The challenge is that AIO provides immediate answers, reducing the incentive to click. The solution is to use Zero-Click spaces to drive conversion actions right on the SERP or directly linked to business goals.

Create a Call to Action (CTA) in Excerpt Content: While the answer should be brief (40-60 words), the copy should contain an implicit or explicit CTA. In how-to articles, it's important to link next steps to a conversion action (e.g. "Download Tan Phat Digital's free Digital Marketing plan template").
Provide High Value Tools: In AI's research mode, users are often looking for practical solutions. Tan Phat Digital can provide high-value resources such as: Digital Marketing plan templates, advertising ROI calculators, or in-depth documents that can only be received with a click. Aligning conversion paths with research needs is key to turning visibility into conversions.

VI.2. Priority Action Framework for Tan Phat Digital

Based on in-depth analysis of AIO, Entity SEO and E-E-A-T, here is the priority strategic action framework for Tan Phat Digital to lead in the AI era:

Strategic Pillars: Content Mastery (AEO)
- Optimization Objectives (LLM) Focus): Switch from Keywords to Answers/Depth/Authority
- Priority Actions for Ton Phat Digital:
  1. Implement Query Fanout: Map comprehensive Digital Marketing topic clusters, build topic depth (Topical Depth).
  2. Invest in Proprietary Data: Publish a "Only" Report Advertising Performance (PPC/Paid Media) Vietnam" quarterly.
  3. Content Structure: Ensure every article has a clear 40-60 word answer and use Tables/Lists for easy extraction.
- References:
Strategy Pillars: P.E.E.A.T Refinement
- Optimization Objective (LLM Focus): Establish Expert/Experience level trust signals
- Priority Actions for Tan Phat Digital:
  1. Reference Documents:

PART VII: FREQUENTLY ASKED QUESTIONS (FAQ) ABOUT AI OVERVIEWS AND SEO

Question: Will AI Overviews kill SEO? Answer: No. AI Overviews (AIO) does not kill SEO but changes its goals. Instead of optimizing for clicks (Traffic), the focus shifts to optimizing for AI citations (Visibility and Authority). More than 99% of AI citations still come from traditional Top 10 results, so maintaining a Top 10 ranking is a prerequisite for survival in the AI era.

Question: How can I increase the chances of my content being cited in AI Overviews? Answer: Focus on Answer Engine Optimization (AEO): 1) Provide direct, succinct answers (40-60 words) right below the headline ; 2) Use Schema Markup (like FAQ and HowTo) to encode a clear answer structure for the AI; 3) Build strong E-E-A-T authority by displaying authors, sources, and proprietary data.

Question: How important is Schema Markup for AI SEO? Answer: Schema Markup is a structural language that helps AI Overviews easily identify, extract, and cite your content. It eliminates the "guessing" for the AI, providing clear signals about what the answer is, who the author is, and when the content was published. Using Schema significantly increases the likelihood of content being selected as a citation source.

Question: Does Zero-Click Search mean there are no conversions? Answer: Not really. While Zero-Click reduces CTR, it creates an opportunity to drive conversions through smart CTAs. Tan Phat Digital should take advantage of this display space to link follow-up actions to valuable resources (e.g. calculators, report templates) that users need to click to download. Users who interact with AIO are often those with higher quality intent.

The future of search has been reshaped by Generative Artificial Intelligence. The shift to AI Overviews is not the end of SEO, but an urgent transition from optimizing for search engines to optimizing for large language models.

To maintain competitiveness and long-term survival, businesses like Tan Phat Digital need to consider the cost of investing in Entity SEO and Proprietary Research as the cost of building a competitive barrier that cannot be copied. By focusing on three strategic pillars—AEO (Answer Optimization), E-E-A-T (Authority & Trust), and Entity SEO— Tan Phat Digital can ensure a leading position in the era of generative search.

The upcoming game is not to trick Google like the old days, but to serve users in the best way so that AI can choose your content as a reference source. Become an AI-recognized authoritative source of information, to convert visibility into brand authority and long-term business value.

Immediate Action: Start with a Query Fanout audit of your current content, and immediately implement Schema Markup on your highest-value pages.