The GEO strategy workbook: A step-by-step guide to getting cited by generative AI

This workbook takes the key ideas from Frase’s guide to Generative Engine Optimization (GEO) and turns them into a practical, repeatable system. You’ll go from understanding how generative engines find content to building your own strategy that gets surfaced, cited, and shared by tools like ChatGPT, Perplexity, and Claude.

Let’s get you GEO-ready.

🛠️ Pro tip: We suggest having a notepad handy as you go through this exercise—this is a workbook after all!

What is the GEO (Generative Engine Optimization) strategy workbook?

This Generative Engine Optimization (GEO) workbook is your comprehensive, hands-on blueprint for adapting your content strategy to the world of AI-driven search. It's an actionable guide designed to help you optimize your digital content to be found, understood, and—most importantly—cited by generative AI models. Think of it as your strategic manual that bridges the gap between traditional SEO and the new demands of answer engines. By working through the assessments, steps, and strategies within, you will build a durable competitive advantage, ensuring your brand's voice and expertise are central to the answers users receive from AI.

Why this GEO workbook matters now

Traditional search engine optimization (SEO) has long been the cornerstone of online strategy, focusing on ranking web pages to attract clicks. But as you're already reading this workbook, you're likely well aware that the landscape has now shifted. And whilst SEO still matters, it's no longer all that matters.

Generative AI powers chat tools like ChatGPT, Perplexity, and Google's AI Overviews. The rise in these tools has drastically changed how people find information. They are no longer just searching; they are asking for answers. In this new landscape, being the source of that answer—being cited—is the new key to success.

This guide is designed as a practical, step-by-step workbook. It will move you from understanding the theory behind this shift to actively implementing a robust strategy.
By following this blueprint, you will learn:

how to audit your existing content
craft new content with AI in mind
implement essential technical signals
measure your success.

You don't need to be a data scientist to get started, just a willingness to adapt your content strategy for the age of AI. The ultimate goal is to transform your content from a passive entry in a list of search results into an authoritative source actively used and cited by the world's most advanced generative engines.

Your Blueprint for Success: What This Workbook Will Achieve

This workbook is designed to be your practical guide to implementing a successful GEO strategy. Upon completion, you will have a clear framework to:

Assess Your Readiness: Audit your existing content to identify strengths, weaknesses, and opportunities for AI citation.
Develop a GEO Content Strategy: Learn to create content that is inherently "citable" by prioritizing clarity, data, and structure.
Implement Technical Optimizations: Master the use of schema markup and other technical signals to communicate explicitly with AI models.
Measure and Iterate: Establish a system for tracking your visibility within AI responses and refining your strategy over time. Ultimately, this blueprint will equip you to build a defensible "citation moat" around your expertise, ensuring your brand remains visible and trusted as AI search becomes the new norm.

Introduction: Navigating the New Era of AI Citation

The digital landscape is undergoing its most significant transformation since the advent of the search engine. We are moving from a world of search results to a world of generated answers. For content creators, marketers, and businesses, this is not just a change. It is a revolution. It needs a new approach, a new plan, and a new way to think about being seen. This introduction sets the stage for Generative Engine Optimization (GEO), the framework for thriving in this new era.

The Shift to Generative Search: Why AI Citation is Paramount

User behavior has fundamentally changed. Instead of sifting through ten blue links on a search engine results page (SERP), users are now engaging in conversations with AI to get direct, synthesized answers. This shift accelerates the trend toward "zero-click searches," where the user's query is answered directly on the results page, eliminating the need to click through to a website. When generative AI like Google's AI Overviews or Perplexity gives a combined answer, the sources it uses gain great authority and visibility. Being cited within these AI-generated responses is the new "ranking." It places your brand not just on the page, but within the answer itself, establishing trust and expertise at the most critical point of the user's journey.

What is Generative Engine Optimization (GEO)? Beyond Traditional SEO

Generative Engine Optimization (GEO) is the practice of creating and optimizing content to be selected and cited as a source by generative AI models in their responses to user queries. While it shares roots with traditional SEO—both value high-quality, relevant content—their primary goals diverge. SEO focuses on achieving a high rank in search results to earn a click. GEO, in contrast, focuses on becoming an embedded, authoritative source within the AI's direct answer. You need to focus more on fact-density, structured data, semantic clarity, and clear expertise. These signals show an AI model that your content is a reliable source for accurate answers.

(There's also AEO to consider which you can read about here)

How does the GEO strategy compare to traditional SEO approaches when targeting generative AI systems?

The fundamental difference lies in the end goal. Traditional SEO optimizes for visibility and clicks from a list of ranked documents. Its tactics revolve around keywords, backlinks, and user experience signals designed to convince a search engine that a specific page is the best destination for a query.

GEO, however, optimizes for attribution and incorporation within a synthesized answer. It's less about being the destination and more about being the trusted source. This means that while keywords are still relevant for topic identification, the emphasis shifts to:

Answer-First Content: Providing concise, direct answers upfront that an AI can easily extract.
Fact-Density: Embedding verifiable statistics, data points, and direct quotes that AI models can use for grounding.
Semantic Structure: Using clear headings, lists, and schema markup to break down information into logical, digestible chunks for machine interpretation.
Authoritative Sourcing: Citing primary sources and linking to other authoritative entities to build a web of trust that the AI can recognize.

How does a GEO strategy help in getting cited by generative AI?

A GEO strategy works by aligning your content with the core processes of generative AI models, particularly Retrieval-Augmented Generation (RAG). The AI retrieves a set of relevant documents. It ranks them for authority and relevance. Then, it uses information from the top sources to generate a new, clear answer. GEO helps you win at each stage of this process. By creating fact-dense, well-structured, and authoritative content, you increase the probability that your pages will be retrieved and ranked highly by the AI. If you provide clear, citable pieces of information like statistics, definitions, or quotes, you help AI use your content in its answers. The AI can also credit you as the source.

Part 1: Understanding the AI Citation Imperative

Before you can optimize for AI citation, you must understand why it's essential and how generative engines "think." This section dives into the mechanics of AI information processing, revealing what makes content trustworthy and citable from a machine's perspective. Mastering these foundational concepts is the first step toward building a successful GEO strategy. (This will also help explain the logic used behind Frase's own GEO scoring system).

Why AI Citation Matters for Visibility and Trust

In the traditional search model, visibility was measured by ranking. In the generative search model, visibility is measured by citation. When your brand is cited in an AI-generated answer, it receives a powerful, implicit endorsement. This placement positions your content not as one of ten options, but as a verified component of the correct answer. This has profound implications for brand perception.

Users are more likely to trust information presented within a synthesized answer, and that trust is transferred to the cited sources. Consistent citation builds your brand's authority by positioning it as a trusted source.

How Generative AI Models Process and Cite Information (LLMs, AI models, generative engines)

Most modern generative engines, from ChatGPT to Google's AI Overviews, rely on a framework called Retrieval-Augmented Generation (RAG). Understanding this process is key to GEO. It unfolds in a few key steps:

Query Understanding: The AI model first analyzes the user's prompt to understand its intent and identify the core entities and concepts.
Information Retrieval: The model then queries a vast index of information (essentially, the internet or a curated dataset) to find documents that are semantically relevant to the user's query. This is similar to a traditional search.
Ranking and Filtering: The retrieved documents are then ranked based on a multitude of signals, including relevance, authority (E-E-A-T), recency, and trustworthiness. The AI selects the top few sources to use.
Answer Synthesis: The Large Language Model (LLM) reads and synthesizes the information from these top-ranked sources, weaving them together to form a coherent, narrative answer.
Citation: Finally, the model attributes the information used in its answer by linking back to the source documents.

Your goal with GEO is to ensure your content is a top contender in steps 2 and 3, making it an indispensable resource for step 4.

The AI's Perspective: What Makes Content "Citable"?

From an AI's perspective, citable content possesses a distinct set of characteristics that signal reliability and utility. These are not just quality guidelines; they are functional requirements for machine processing. The most citable content is:

Verifiable: It contains facts, statistics, and quotes that can be cross-referenced with other authoritative sources.
Structured: It uses clear headings (H2s, H3s), bullet points, and numbered lists to organize information into logical, discrete chunks. This makes it easier for the AI to parse and extract specific pieces of information.
Concise: It provides direct, unambiguous answers to specific questions without unnecessary fluff. An "answer-first" approach is highly effective.
Authoritative: It demonstrates expertise through comprehensive coverage, citing primary research, and linking to other well-regarded sources.
Recent: For many topics, fresh and up-to-date information is prioritized, signalled by publication and update dates.

Grounding and Fact-Density: The Pillars of AI Trust Signals

"Grounding" is the process by which an AI model anchors its generated statements in verifiable facts from its source documents. This is a critical mechanism for preventing "hallucinations" or the creation of false information. Your content can become a prime grounding source by increasing its fact-density. This means systematically including hard data, statistics, named entities, dates, and specific figures throughout your text. Content rich with verifiable facts provides the raw material the AI needs to build a trustworthy answer. Aim to support your main points with specific data wherever possible, as this makes your content not just informative for humans but functional for AI.

Building E-E-A-T (Experience, Expertise, Authoritativeness, Trustworthiness) for Generative Answers

Google's E-E-A-T framework, a cornerstone of modern SEO, is even more critical for GEO. Generative models are explicitly trained to identify and prioritize sources that demonstrate these qualities.

Experience: Showcase firsthand knowledge. Include case studies, original research, or content written by practitioners with real-world experience.
Expertise: Go deep on your topic. Create comprehensive content that covers a subject from all angles, answering niche questions and demonstrating a command of the subject matter.
Authoritativeness: Build your reputation both on and off your site. Earn links and mentions from other respected sites in your industry. Secure author bylines on reputable publications.
Trustworthiness: Be transparent. Clearly state your sources, provide author bios with credentials, and make contact information easy to find. Secure your site with HTTPS.

For generative AI, E-E-A-T isn't just a ranking factor; it's a core filter for selecting reliable source material.

Part 2: Your AI Citation Readiness Assessment (Workbook Exercise)

This section transitions from theory to practice. Before building your GEO strategy, you need a clear picture of where you currently stand. These exercises help you do a full self-check. They set a starting point to measure progress and find your best chances to improve.

Introducing the GEO Citation Readiness Scorecard (exercise, notepad)

Grab a notepad or open a new document. You're going to create a simple scorecard to assess a key piece of your content. Choose an important article you want to optimize. For each of the following criteria, give it a score from 1 (Poor) to 5 (Excellent).

Answer-First Clarity: Does the article provide a direct, concise summary (40-60 words) of the main topic near the beginning?
Fact-Density: How many verifiable statistics or data points does the article contain per 500 words? (1-2 = Poor, 5+ = Excellent)
Structured Data: Does the page use schema markup (e.g., Article, FAQPage)?
Clear Organization: Is the content broken down with clear, descriptive subheadings (H2, H3)?
Authoritative Sourcing: Does the article cite and link to primary research or recognized authorities?
Author Expertise: Is the author clearly identified with a bio and credentials?

Total your score. A score below 18 indicates significant room for improvement, while a score above 25 suggests a strong foundation for GEO.

Auditing Existing Content for Citatability and Attribution Gaps

Now, expand the scorecard exercise into a broader audit.

Create a spreadsheet listing your top 10-20 most important content assets. For each URL, create columns for the scorecard criteria above. Systematically go through each piece and score it.

This process will quickly reveal patterns. You might find your content is well-organized but lacks fact-density, or that you consistently forget to implement schema markup. The most important column to add is "Action Items."

For each low-scoring piece of content, note the specific steps needed to improve its citatability (e.g., "Add 3 new statistics," "Implement FAQ schema," "Rewrite intro to be answer-first").

Identifying Content Gaps & Opportunities for AI Keywords and Answer Engine Optimization

Your audit reveals how well your existing content is optimized. The next step is to find what's missing.

Use tools like AnswerThePublic, AlsoAsked, or simply analyze the "People Also Ask" sections in Google search results for your core topics (top tip: you can utilise the optimization tab in Frase for this). Look for recurring questions that you haven't answered comprehensively. These questions are direct inputs into AI search engines. Creating dedicated content or content sections that directly answer these queries is a primary GEO tactic. Think less about traditional high-volume keywords and more about the conversational questions and complex problems users are trying to solve with AI.

Competitor Comparisons: Learning from AI-Cited Rivals (competitor comparisons)

Identify a few key queries in your industry and pose them to different AI engines (ChatGPT, Perplexity, Google AI Overviews). Note which competitors are being cited in the responses. Analyze their cited pages using your scorecard criteria.

How is their content structured?
What kinds of data or quotes are being pulled?
What schema markup are they using?
How do they establish their E-E-A-T? This reverse-engineering process is invaluable. You're not looking to copy your competitors, but to understand the specific content attributes that are currently being rewarded by AI models in your niche. This analysis provides a clear roadmap for what success looks like.

Setting Baseline Metrics for AI Visibility (AI Visibility Tracking, Geostar Visibility Tracker)

You can't improve what you don't measure. Before you begin implementing your GEO strategy, you must establish a baseline. The most direct method is manual tracking.

Create a list of 20-30 critical, question-based queries relevant to your business.
Run these queries through your target AI engines (e.g., ChatGPT, Perplexity) on a set schedule (e.g., the first of every month).
In a spreadsheet, log whether your domain was cited for each query, and if so, what page was referenced. This creates a "GEO Visibility Score" (e.g., "Cited in 3 out of 30 queries = 10% visibility"). While tools for automated tracking are emerging, this manual process provides granular, invaluable insight into your current performance and will allow you to demonstrate the ROI of your GEO efforts over time.

Part 3: Step-by-Step Blueprint for Proactive AI Citation

With your audit complete and baseline established, it's time to build. This section provides a five-step blueprint for creating and optimizing content that is designed from the ground up to be cited by generative AI. Follow these steps to systematically enhance your content's appeal to AI models.

Step 1: Foundational Content Strategy for Inherent Citatability

Your journey begins with strategy. Shift your content planning from a purely keyword-driven model to a topic-and-question-driven one. Focus on creating comprehensive "hub" style pages (Pro-tip: you can use Frase's Rank-Ready Documents or AI Content Generation features to build your hub pages from scratch). These pages cover a topic completely. They answer the main questions and related secondary and tertiary questions. Aim to become the definitive resource on a subject. This approach builds topical authority, a powerful signal for AI engines. Prioritize content that addresses complex, multi-step queries—the kinds of prompts users bring to generative AI. Your goal is to create the go-to resource that an AI would need to consult to form a complete and accurate answer.

Step 2: Crafting Content for AI Attribution and Fact-Density

This step focuses on the micro-level details of your writing.

Adopt an "Answer-First" Structure: Begin your articles with a concise, 40-60 word summary that directly answers the central question of the piece. This is prime material for extraction by AI.
Increase Fact-Density: Methodically weave verifiable data throughout your content. For every major point you make, ask, "Is there a statistic or data point I can add to support this?" Cite your sources clearly. Research has shown that specific GEO strategies, like including statistics and citations, can increase citation frequency by up to 40%.
Use Question-Based Headings: Structure your articles with H2 and H3 headings that mirror the questions users are asking. This makes your content easy to scan for both humans and AI crawlers.
Incorporate Direct Quotes: Include quotes from industry experts, primary sources, or internal stakeholders. Formatting these with blockquotes makes them easily identifiable and citable units of content for an AI.

Step 3: Advanced Structured Data & Schema for Explicit AI Citation

Schema markup is a form of microdata that provides explicit context to search engines about your content. It is a direct communication channel to AI. Prioritize implementing the following types:

Article Schema: Defines your content as an article, specifying the author, publication date, and headline.
FAQPage Schema: Use this for pages with a question-and-answer format. Mark up each question and its corresponding answer to make them easily extractable for AI-powered answer snippets and responses.
Person Schema: On your author pages, use Person schema to detail their credentials, affiliations, and areas of expertise, directly bolstering your E-E-A-T signals.
Organization Schema: Ensure your site's main entity is clearly defined with Organization schema, linking to your social profiles and establishing your brand as a recognized entity.

Step 4: Technical Optimization for AI Content Retrievability

Your brilliant, fact-dense content is useless if AI models can't efficiently access and process it. Ensure your technical foundation is solid.

Crawlability and Indexability: Ensure your site has a clean XML sitemap and a logical internal linking structure. Use robots.txt correctly to avoid blocking important content from crawlers.
Page Speed: Fast-loading pages are preferred. AI models operate on a massive scale and favor content that can be retrieved and processed quickly.
Mobile-First Design: Since many AI queries originate on mobile devices and models often use mobile crawlers, a responsive and clean mobile experience is non-negotiable.
Clean HTML: Use semantic HTML5 tags (e.g., <article>, <section>, <aside>) to provide additional structural clues about your content's hierarchy and meaning.

Step 5: Amplifying Your Semantic Footprint and Authority

GEO is not just about on-page content; it's also about how your content is perceived within the broader digital ecosystem.

Build Topical Clusters: Create a network of interlinked articles around a central "hub" topic. This demonstrates a deep level of expertise and helps AI models understand the relationships between different concepts on your site.
Strategic External Linking: Link out to authoritative primary sources, academic studies, and respected industry publications. This shows that your content is well-researched and positions it within a constellation of trustworthy information.
Earn Authoritative Mentions: While traditional link building is still valuable, mentions of your brand, authors, and research on other authoritative sites—even without a hyperlink—contribute to your entity recognition and overall authority in the eyes of AI.

Part 4: Measuring, Monitoring, and Iterating Your AI Citation Strategy

Implementing a GEO strategy is not a one-time task; it's an ongoing process of optimization. To succeed, you need a robust system for measuring performance, analyzing results, and iterating on your approach. This final section provides a framework for tracking your GEO success and ensuring your efforts deliver tangible results.

Defining GEO-Specific KPIs for AI Citation Success

Traditional SEO KPIs like rank position and organic clicks, while still relevant, don't tell the whole story for GEO. You need to adopt new, specific Key Performance Indicators (KPIs):

Citation Frequency: The percentage of target queries for which your domain is cited. This is your primary GEO visibility metric.
Citation Share of Voice: Your citation frequency relative to your key competitors for a shared set of target queries.
Quality of Citation: Are you being cited for high-level, definitional queries or for more nuanced, high-intent questions?
Attribution Accuracy: When cited, does the AI accurately represent the information from your content?
Click-Through Rate from Citations: For AI engines that provide clickable citations, what is the traffic coming from these sources? (This can be tracked using UTM parameters where possible or by analyzing referral traffic).

Tools for Performance Tracking and Analysis

While the GEO toolkit is still evolving, you can combine several existing resources for effective tracking:

Manual Tracking Spreadsheet: As established in Part 2, this is your most reliable tool. Consistently run your target queries through AI engines like ChatGPT, Perplexity, and Google and log the results.
Google Analytics 4 (GA4): While it's difficult to isolate AI-driven traffic perfectly, you can create reports to monitor referral traffic from domains associated with AI search (e.g., perplexity.ai). You can also look for spikes in direct traffic to specific, highly-citable URLs after you've observed a citation.
Google Search Console (GSC): Monitor performance in Google's AI Overviews. While GSC doesn't yet separate this traffic perfectly, analyzing clicks and impressions for long-tail, question-based queries can provide directional insights.

Analyzing AI Responses and AI Overviews for Citation Accuracy

Monitoring isn't just about counting citations; it's about analyzing their quality. When you find a citation, ask critical questions:

Is the context correct? Did the AI use your information in the way you intended?
Is the snippet accurate? If a specific quote or data point was pulled, is it free from misinterpretation?
What content was used? Was it the "answer-first" summary, a specific statistic, or a quote from an expert? Understanding what is being extracted helps you refine your content creation process. If you find inaccuracies, it's a valuable feedback loop. It may indicate that your content lacks clarity or that a specific section could be rephrased to be less ambiguous for machine interpretation. Continuously analyzing these responses is key to long-term iteration and improvement.

Next Steps After Completing Generative Engine Optimization Workbook

You have now completed the Generative Engine Optimization Workbook. You have moved from understanding the big change in search to checking your current content. You are making a plan to improve and setting up ways to measure progress. You are no longer a passive observer of the AI revolution in search; you are an active participant equipped with a clear strategy.

The key takeaway is that Generative Engine Optimization is not a replacement for SEO but a critical evolution of it. The goal has shifted from simply ranking to becoming a trusted, foundational source of information for AI models. When you create fact-dense, well-structured, and trustworthy content, you build a "citation moat." This is a lasting advantage that others find hard to copy.

Your next steps are clear and actionable:

Implement Your Audit Findings: Begin with the low-hanging fruit identified in your content audit. Update your most important articles with answer-first summaries, additional statistics, and proper schema markup.
Integrate GEO into Your Workflow: Ensure the five-step blueprint from Part 3 becomes a standard part of your content creation process for all new articles.
Commit to Consistent Tracking: Start your manual citation tracking immediately. The baseline data you collect now will be invaluable for demonstrating progress and refining your strategy in the months to come.

The world of AI search will continue to evolve, but the principles of expertise, authority, and trust are timeless. By using GEO, you do more than optimize for today's algorithms. You protect your content strategy for the future. You also make your brand a trusted source for the next generation of search.