Whether You Call it SEO, AEO, or GEO, the E-E-A-T Model is Insufficient

51% of B2B software buyers now start their research with an AI chatbot versus Google. This is according to G2’s latest research, from April 2026, up from 29% the previous year.

Meanwhile, Google guidance stays true to its narrative that it’s all just SEO and the fundamentals are unchanged.

If this were true, we’d see much greater consistency between the pages that rank in traditional search results and the citations and recommendations presented on AI search. This study by Ahrefs found that only 12% of AI citations overlap with Google’s traditional top 10 search results. 

Clearly, the underlying weights and measures are different. And that’s why we need a new optimization framework.

A Brief History of E-E-A-T

While it was created in 2014, E-A-T — Expertise, Authoritativeness, Trustworthiness — did not explode in popularity within the SEO community until 2018.

The August 2022 Helpful Content Update added an E — Experience — and cemented E-E-A-T as the defining blueprint for modern SEO across most categories. Experience was added to the original model in large part to reward human-generated content in the era of AI.

By embedding these signals directly into its core ranking systems, Google forced the industry away from short-term algorithmic tricks and back to a system that attempted to reward content that was helpful and high quality.

Here we are now, four years into commercially available LLM-powered search. Is E-E-A-T still the best framework to help brands and content creators maximize discoverability and brand awareness?

We Need a New Search Optimization Model.

AI search doesn’t rank pages; it builds answers. SEO optimizes for documents; AI optimizes for knowledge objects.

That’s not insignificant.

Traditional search ranked content and asked users to sort through it. AI search scans sources, extracts relevant information, and assembles a response around the specific context of the prompt.

As a result, people approach search differently now.

They write long-form prompts with context, constraints, goals, preferences, and background information. The more information we provide upfront, the more specific and contextual the response is. 

Try that with a traditional search query and you’ll likely get increased unhelpful variance and decreased relevance in the search results. 

E-E-A-T was built for ranking pages in traditional search, with a bolt-on extra E to address AI-generated commodity content. That happened when AI assistants accounted for less than 2% of total search activity. Now that number is estimated to be around 50%.

An extra E doesn’t cut it. We need a new framework for content creators that optimizes for both AI and traditional SERP placements.

Introducing the C-U-R-A-T-E Framework.

E-E-A-T tells search engines how to evaluate content. C-U-R-A-T-E tells brands how to create content that optimizes for both AI and traditional search. You’ll notice some overlap.

Letter	Principle	Definition
C	Context	Fit the buyer’s specific situation.
U	Uniqueness	Contribute something new that AI can’t generate by itself.
R	Relevance	Answer the right question for the right buyer at the right moment.
A	Alignment	Reinforce the same positioning across touchpoints.
T	Trust	Demonstrate credibility through people, proof, and transparency.
E	Experience	Share first-hand knowledge that only humans possess.

Let’s unpack each of these 6 signals one by one.

 

C — Context. Fit the Buyer’s Specific Situation.

Traditional SEO optimized around keywords and topics. AI search optimizes around situations.

In the world of traditional SEO, too much context often led to scattered results. In AI search, the more context the better. Information about the desired goal, application, use case, constraints, output format, industry, or buying stage helps the model synthesize a better answer.

Large language models are pattern-recognition systems. Traditional authority signals may still influence what they consider, but their higher-order priority is finding information that best fits the user’s situation. Context carries more weight.

Take Ally Bank.

Historically, a search engine might have favored the overall authority of a giant institution like Bank of America.

But if someone asks:

“What’s the best online bank for high-yield savings and low fees?”

Ally suddenly has a chance to become the more relevant answer.

That’s a shift from broad authority to contextual authority.

The implication for marketers is significant. Your job is no longer just understanding the keywords your buyers search. It’s understanding the situations they find themselves in.

Build a Prompt Inventory, Not a Keyword List.

Most marketing teams still organize content around keywords. But buyers don’t naturally think in keywords. They think in situations, constraints, risks, tradeoffs, and desired outcomes. AI simply gives them a better way to express those thoughts.

That’s why succeeding at AI search demands a deeper understanding of the questions behind the keywords.

One exercise we run with clients is building a Prompt Inventory. Think of it as the evolution of keyword research.

Instead of asking:

“What keywords should we rank for?”

Ask:

“What situations does our buyer find themselves in, and what would they ask AI in that moment?”

Here are four ways to build that inventory.

1. Existing Keyword Data

Take every keyword and ask:

What problem is the buyer actually trying to solve?

For example:

“ERP implementation challenges”

becomes

“How can I replace spreadsheets without disrupting operations?”

Or:

“Marketing automation benefits”

becomes

“How can I scale lead nurturing without hiring another marketer?”

The keyword describes a category. The prompt reveals the core customer problem.

2. Communities

Look where your buyers already discuss their problems.

That might be:

• Reddit
• LinkedIn Groups
• Slack communities
• Industry forums
• Product review sites

The language people use there often sounds much closer to an AI prompt than anything you’ll find in a keyword research tool.

3. Voice of the Customer

Sales and customer service conversations are a goldmine of context.

Every question asked during:

• discovery calls
• demos
• trade shows
• customer interviews
• procurement conversations

…is a future AI prompt.

4. AI Itself

Ask AI to generate a hundred or more prompts a buyer might ask while evaluating solutions in your category.

Then cluster them by:

• problem
• persona
• buying stage

The result isn’t just a keyword universe. It’s a buyer situation map.

The brands that understand those situations best—and consistently create content around them—are the ones most likely to become part of the answer.

U — Uniqueness. Contribute Something New.

Large language models already know the internet’s common knowledge. When your content simply restates what the model already knows, it has very little reason to cite you. To become part of the answer, your content needs to contribute something new.

The Goal is Information Gain.

Every article, podcast, case study, framework, or customer story should teach the model something it didn’t already know.

Recent research by Flying V Group illustrates this well. When a page simply restates information already contained within the model’s knowledge, the model answers from memory and skips the citation. This helps explain why so much traditional SEO content is losing ground in AI search.

For years, SEOs were encouraged to create “Round ups” and “What is…” and “How to…” articles. Those content types worked when Google was purely in the business of directing traffic and not answering questions. Now, over half of all searches are “zero-click”. 

The same Flying V study found that less than 8% of these commoditized content types generated meaningful AI traffic. The winners are the brands creating knowledge that the model cannot easily infer from what it already knows.

That includes:

• Proprietary frameworks
• Original research
• Customer stories
• Lessons learned
• Failed experiments
• Contrarian viewpoints
• Unique methodologies
• First-hand observations

Generative AI struggles to create these because they originate with you.

What About Trained AI Agents?

The aspiration of custom agents that know the brand deeply enough to produce differentiated marketing content is possible. It’s just more difficult than most companies realize. And it must be built on top of foundational brand strategy, human-led thought leadership, and customer use-cases as identified through VoC and other human-originated sources.

Custom agents are only as differentiated as the knowledge they’re trained on. If the source material is generic, the output will be generic.

If the source material contains original thinking, customer insight, lived experience, and proprietary frameworks, AI becomes an amplifier.

 

R — Relevance. Answer the Right Question for the Right Buyer at the Right Moment.

Context determines what someone is trying to accomplish.

Relevance determines whether your content is the right answer for that person at that moment.

When a brand consistently speaks to a specific audience, several things begin to align: Customer reviews reinforce the same strengths. Case studies tell similar stories. Third-party conversations describe the brand in consistent ways.

Over time, AI begins to associate your brand with solving a particular type of problem for a particular type of buyer.

That’s what happens when someone asks:

“What’s the best ERP strategy for a mid-sized manufacturing company that needs to improve inventory accuracy without disrupting operations?”

Now compare that with:

“Give me an executive comparison of NetSuite and SAP for a CFO.”

Both prompts are about ERP systems.

But they represent different buyers with different needs at different stages of the buying journey. And the content required to answer them is completely different.

Build a Content Matrix, Not Just a Content Calendar.

Most content strategies are organized by topics. The best AI content strategies are organized by buyer situations.

One exercise we use with clients is creating a simple Content Matrix.

Across one axis:

• ICPs and/or Personas

Across the other:

• Buying stages

Each intersection represents a unique conversation. For each buyer persona at each buying stage, you should know the problems they face, their mindset, what they need, as well as have a distinct key takeaway, supporting soundbytes, and offer / CTA.

A CFO evaluating ERP vendors has different questions than an Operations Director exploring inventory challenges. A prospect discovering they have a problem needs different content than one comparing vendors. Yet many brands try to answer all of those questions with a single narrative.

That’s no longer enough.

Yes, this creates more narratives. But it also creates far greater relevance, which increases engagement and conversion throughout your marketing and sales lifecycle.

A — Alignment. Reinforce Consistent Brand Identity.

Section Preface: Everything we talk about here that is true for humans is also true for AI, and vice versa. 

Even content with good context, uniqueness, and relevance may not be enough to break through. You also need alignment with a core brand identity or positioning.

Now, this might seem at odds with everything we described thus far. How do you be hyper-contextual and relevant to specific use-cases and buyer journey stages, AND be consistent? 

This is an age-old dilemma. But the short answer is, the goal of the brand architecture is to establish unifying principles that anchor your core identity, ensuring your content remains recognizable even as it adapts to the specific needs of diverse buyer contexts.

Consumers develop a mental “schema” for a brand; if that schema is constantly contradicted by inconsistent messaging, the consumer loses the ability to categorize the brand, leading to lower recall and reduced trust. The same holds true for AI models. 

Humans Infer Your Brand. So Does AI.

Large language models don’t evaluate a single page in isolation. They build an understanding of your brand by recognizing patterns across everything they can access:

• Your website
• Your LinkedIn posts
• Your podcast
• Customer reviews
• Case studies
• Press coverage
• Discussion forums
• Videos
• Employee content
• Executive thought leadership

Every one of these becomes a signal that helps AI answer a simple question:

“What is this brand actually known for?”

The easy mistake is viewing these individual marketing assets in isolation, which leads to fragmentation, noise, and distrust.

Optimize Your Signal-to-Noise Ratio.

In a noisy market, just like in a noisy auditorium, a single voice won’t cut through. But many voices can when they are coordinated. The brand positioning, clearly and consistently expressed throughout all your touchpoints, acts like a beacon, for both humans and large language models scanning for content-context fit.

Every touchpoint either adds signal or noise. The imperative becomes clear when you realize how much repetition it takes to drive brand consideration in the B2B context.

According to Forrester, modern B2B buyers average 27 interactions before initiating contact. In complex industries, RAIN Group estimates buyers experience 40–50 touchpoints before making a decision.

Alignment doesn’t mean every asset says the same thing. Your messaging should absolutely adapt to different audiences, different use cases, and different stages of the buying journey.

Underneath those variations, however, should be a recognizable identity—a consistent point of view, a clear value proposition, and a set of core ideas that your brand becomes known for.

Every interaction should make your brand more recognizable.

The stronger and more consistent the signal, the easier it becomes for both buyers and AI systems to understand what your brand stands for and when it should be recommended.

T — Trust. Demonstrate Credibility Through People, Proof, and Transparency.

Trust has always been important in marketing. What’s changing is how it’s earned.

Brands used to have much greater control over their own narrative. If a company claimed to be innovative, customer-focused, or the market leader, buyers often had little choice but to evaluate those claims on the company’s own terms.

Today, both buyers and AI have far more ways to verify them.

Trust Is Inferred, Not Declared.

Large language models don’t simply repeat what your website says. They compare your messaging against customer reviews, case studies, industry publications, executive thought leadership, employee perspectives, and countless other third-party signals before determining whether your claims are credible.

In other words, trust is no longer something a brand declares. It’s something that buyers—and increasingly AI—infer from the weight of available evidence.

This mirrors how people make decisions. We rarely trust a company because it says it’s trustworthy. We trust it because we encounter enough consistent evidence to believe the claim.

Let Others Build Your Credibility.

The strongest trust signals rarely originate with the marketing department. They come from customers sharing their experiences, employees demonstrating expertise, partners advocating on your behalf, analysts covering your industry, and respected publications validating your work.

That’s why customer stories, reviews, testimonials, podcast appearances, conference presentations, PR coverage, and executive thought leadership have become increasingly valuable. They aren’t simply marketing assets—they’re independent trust signals that corroborate your brand’s claims.

Humans Trust Humans.

Nearly half (43%) of digital users report no longer trusting most online content due to the sheer volume of synthetic media flooding the internet. 

Content Category	Trust AI-Generated More	Trust Human-Generated More
News & Current Affairs	8%	68%
Product Reviews & Recommendations	9%	66%
Legal Documents & Contracts	10%	66%
Social Media Posts & Articles	7%	69%

Source: Pangram AI Sentiment Survey

We’ve seen Reddit and other UGC platforms, as well as influencers and thought leaders, become more influential than brands in recent years since the arrival of AI. This is true both in terms of human influence and LLM influence. 

B2B brands must activate individual thought leadership and brand advocacy through its senior leaders and subject matter experts.

E — Experience. Capture and Share First-Hand Expertise.

One question you may have noticed is what happened to Expertise from Google’s E-E-A-T framework.

Google originally separated Experience and Expertise because they serve different purposes in evaluating content. Experience asks, “Have you actually done this?” Expertise asks, “Do you know what you’re talking about?”

In practice, however, those two ideas are deeply connected. The most compelling expertise is rarely theoretical. It comes from solving real problems, working with real customers, making mistakes, refining approaches, and accumulating practical knowledge over time.

That distinction becomes even more important in an AI-first world.

Large language models are remarkably good at synthesizing existing knowledge. What they can’t do is manufacture genuine experience. They can’t interview your customers, lead a difficult implementation, learn from a failed product launch, or develop a proprietary methodology over years of working with clients. Those experiences become expertise—and they remain one of the most durable competitive advantages a brand can have.

Every Expert Is a Content Asset.

One of the biggest untapped opportunities inside most organizations is the expertise that never gets documented. Your sales team understands customer objections. Your consultants know where implementations succeed or fail. Your customer success team sees the patterns behind long-term adoption. Your executives have perspectives shaped by years of strategic decisions.

Every one of those people possesses knowledge that AI cannot invent because it originates from lived experience. You already have the expertise. The challenge is creating systems that consistently capture it.

Scale Expertise, Not Content.

Many organizations have embraced AI to produce more content. A more effective strategy is to use AI to scale human expertise.

Interview your subject matter experts. Record customer conversations. Document implementation lessons. Turn webinars into articles. Repurpose podcasts into research summaries. Extract insights from project retrospectives. Use AI to organize, refine, and distribute that knowledge, but make sure the ideas originate with people who have actually lived them.

The goal isn’t to publish more. It’s to publish more of what only your organization knows. That’s the kind of experience that builds trust, differentiates your brand, and gives AI something genuinely valuable to recommend.

Final Thought: Build a Stronger Signal and a Stronger Brand.

If there’s one theme running through this framework, it’s that AI search is rewarding many of the same disciplines that have always built strong brands.

Traditional search relied heavily on proxies. Because it couldn’t fully understand intent, context, credibility, or expertise, it depended on signals like backlinks, domain authority, keyword relevance, and technical optimization to estimate quality.

One definition of a brand is “the intangible sum of a product’s attributes: its name, packaging, and price, its history, its reputation, and the way it’s advertised” (Ogilvy). Others say a brand lives in the mind of the consumer, and is the sum of all perceptions and attitudes toward the company or product. 

Large language models are increasingly able to evaluate brands the way people do. They look for context, originality, consistency, credibility, and experience. They assemble an understanding of your business from thousands of signals across your website, content, customer stories, reviews, thought leadership, employees, and third-party conversations.

The role of marketing shifts from creating rankable content to creating a coherent body of knowledge that helps both buyers and AI understand who you are, what you stand for, and when you’re the right choice.

Your brand identity defines what you want to be known for and creates alignment and consistency.

Your content strategy drives context and relevance that adapts that identity for different buyers, different situations, and different stages of the buying journey.

Your subject matter experts leverage unique experience in the creation of content assets that bring a unique and trustworthy point of view to the world.

Tactically speaking, when those layers work together, every touchpoint becomes another reinforcing signal.

The C-U-R-A-T-E model helps brands move beyond commoditization, tap the authenticity people crave, and position themselves for the AI frontier.