I embarked on a journey to uncover whether AI crawlers favored Markdown over HTML. By conducting a controlled experiment, I aimed to see if serving content in Markdown format would result in increased bot traffic. After analyzing data from 381 pages over the span of three weeks, I’m eager to share what I discovered.
The results of this experiment could provide valuable insights for those interested in enhancing the visibility of their content through strategic formatting. Stay tuned as I reveal the intriguing findings.
I’ve discovered how essential it is to integrate trusted search intelligence across our enterprise. With the Conductor Data API, we’re extending these capabilities in ways I hadn’t imagined.
Seeing our data work harmoniously across platforms feels transformative, allowing us to leverage AI infrastructure like never before. This powerful insight has reshaped how we view our enterprise integration strategies.
I’ve learned that SEO is evolving beyond just marketing and into the realm of organizational design. It’s about structuring, validating, and aligning information across the business to enhance visibility.
When our information becomes fragmented, visibility suffers, leading to more than just rank instability; it threatens our brand’s interpretation and mentions.
For those of us leading SEO, the choice is clear: remain as channel optimizers or become architects of systems that define brand understanding and citation. With AI systems now assembling information at scale, this transformation isn’t happening in isolation.
The visibility shift beyond rankings
As we look to the future, LLMs will shape organic search alongside traditional algorithms. Simply chasing rankings isn’t enough. We need to optimize for interpretation, citation, and synthesis across AI systems.
While our click rates may vary, the real transformation means treating visibility as an interpretation issue rather than just positioning. AI systems rely on cohesive data, narratives, and mentions. Discrepancies lead to inconsistent output.
Now, collaboration can’t be casual. LLMs demand clarity and consistency in the information they process. When our messages and data are fragmented, so too is our visibility.
This is a leadership challenge. Our visibility shouldn’t exist in silos but as a system that manages information creation, validation, and distribution across our organization.
If we want structural visibility, we must build the system to support it.
Building the visibility supply chain
I’ve realized that collaboration needs to be ingrained in the supply chain. It shouldn’t rely on the relationship between the SEO manager and PR manager.
For a seamless transition from a marketing silo to an operational framework, we must treat content as a product needing precise refinement before entering the broader ecosystem.
Enter visibility gates—nonnegotiable checkpoints filtering brand data for AI consumption.
Implementing visibility gates
Think of our content moving through a pipeline. At each joint, a gate refines the output:
The technical gate (parsing)
Does the product page template use valid schema.org markup (product, FAQ, review)? The goal here is to ensure that the data is structured for seamless AI consumption.
The brand signal gate (clustering)
Does our PR align with core entities? Are we using consistent terminology to help LLMs cluster our brand? This phase aims to prevent linguistic drift.
The accessibility/readability gate (chunking)
Is the content ready for RAG systems? Here, we focus on delivering high-information-density prose suitable for AI retrieval.
The authority and de-duplication gate (governance)
Does this asset induce “knowledge cannibalization” or unnecessary noise? This step filters out conflicting information to ensure a single source of truth.
The localization gate (verification)
Is entity information consistent globally? This ensures alignment in cross-referencing data to build trust in models.
While gates guard the ecosystem entry points, accountability is key to ensure changes translate into actions.
Embedding visibility into cross-functional OKRs
Alignment without visible results can’t sustain change. Sophisticated infrastructure will fail without cross-functional influence.
To advance beyond mere collaboration, visibility must be embedded into our organizational performance strategy.
We need shared visibility OKRs, not just SEO-specific goals.
When stakeholders are incentivized with visibility KPIs, SEO becomes more than just a team job. It becomes a vital business priority.
For product teams: “Achieve 100% schema validation and <100ms time-to-first-byte for top-tier entity pages.”
For PR and communications: “Increase ‘brand-as-a-source’ citations in LLM responses by 15%.”
For content teams: “Ensure 90% of new assets meet ‘high information density’ for RAG retrieval.”
This collective focus aligns our organization with modern search engine mechanisms.
Measuring visibility across the organization
While gates assess the quality entering our ecosystem, a unified dashboard measures output, enhancing transparency.
If PR teams understand which mentions drive AI citations, they’ll focus on authoritative publications rather than any media outlet.
We need to transition from rank reporting to assessing entity health and Share of Model (SoM). This dashboard becomes our brand’s single truth source.
While systems and incentives are important, they need active management.
Join us at Semrush for tools to enhance your brand’s visibility.
Hiring for AI-era visibility
Building a visibility system isn’t enough; we need a workforce matching this model. We should hire beyond generalists, focusing on two pillars: the hacker and the convincer.
Feature
The hacker (technical architect)
The convincer (visibility advocate)
Core mission
Ensuring the brand is discoverable by machines.
Ensuring the brand is supported by humans.
Primary domain
RAG architecture, schema, and LLM testing.
Cross-departmental OKRs, C-suite buy-in, and PR alignment.
Success metric
Share of model (SoM) and information density.
Resource allocation and budget growth.
The gate focus
Technical, accessibility, and authority gates.
Brand signal and localization gates.
The hacker: The engine room
The technical visionary, constantly pushing boundaries. They optimize beyond just search bars, ensuring our brand’s discoverability by AI.
The convincer: The social butterfly of data
This individual bridges technical insights with business results, ensuring the hacker’s ideas are realized in practice.
As we adapt to operational SEO, consider these first steps:
Set the vision: Define what visibility-first looks like for your business.
Take stock of talent: Do you have the hackers and convincers? Evaluate for skills and mindset.
Audit the gaps: Identify and address communication breakdowns between SEO and PR, or SEO and product teams.
Shift the KPIs: Focus on authority, impressions, sentiment share, and revenue, not just rankings.
Be radically transparent: Share data in real time, without siloed thinking.
During the first 90 days, focus on:
Days 1-30 (Audit): Map your brand’s entity footprint to identify conflicts.
Days 31-60 (Infrastructure): Integrate visibility gates into your CMS or project management tools like Jira or Asana.
Days 61-90 (Incentives): Link 10% of PR and product teams’ bonuses to entity integrity or citation growth in AI.
The SEO leader as a systems architect
As AI advances, an effective SEO leader transitions to becoming a systems architect, crafting the infrastructure for both machine and human brand interactions.
This journey is complex, demanding upfront changes to ingrained processes and transparent communication.
The future goes beyond keywords to optimizing how information flows through the digital ecosystem, embracing this transition will build a resilient organization visible by default.
Agentic AI is now a hot topic among executives. I’m here to break down precisely what’s happening, what remains unchanged, and how e-commerce brands should adapt.
As an SEO leader working with e-commerce brands, I’m often in the position of clarifying the realities behind buzzwords like ‘agentic AI’. Executives frequently inquire about its implications for growth, risk, and competition.
Executives crave facts over hype. They seek concise explanations, grounded insights, and actionable advice.
My role as an SEO leader becomes essential here, not in predicting the future, but in enlightening leadership about the changes, the constants, and how to proceed pragmatically. Here’s my roadmap.
Start with Defining ‘Agentic’
First, I focus on demystifying the term. Agentic systems don’t replace customers; they work on their behalf. While the intent and preferences originate from individuals, the execution is taken over by the software.
The working dynamics shift, where tasks like discovery, comparison, and even execution are now managed by software, processing data faster than any human.
In discussions with executive teams, I emphasize simple illustrations:
“We’re not losing customers; instead, we’re incorporating a new decision-maker, which is the software acting as a customer proxy.”
Understanding this calms the conversation and steers focus away from fear towards preparation.
Manage Expectations to Avoid Hype
Another key role I play is in tempering expectations. Agentic AI won’t sweep over all at once. Its effects will be gradual and varied across different categories.
Some industries, with standardized products and organized data, will adapt faster. Others will face more challenges due to complexities and regulatory hurdles.
I often see leadership teams falling into two detrimental traps:
Panic: Hastily altering strategies and budgets without clarity.
Dismissal: Ignoring changes until it impacts performance, leading to rushed responses.
I offer a steady perspective, noting that agentic AI merely accelerates existing trends. It’s not about chasing new features but reinforcing strong fundamentals.
I encourage conversations to evolve beyond search rankings. When agents lead the journey, the critical question becomes, “Are we eligible to be chosen?”
Eligibility hinges on clear, consistent, and trustworthy data. Agents must grasp your offerings, target audience, pricing, availability, and risk factors associated with choosing your brand.
Raising thoughts about data consistency, pricing reliability, and whether policies add or reduce uncertainty positions SEO as a practical bridge between strategy and execution.
SEO Beyond Marketing
There’s a misconception that SEO is confined to marketing. Agentic behavior challenges this notion.
Selection by an agent involves variables beyond marketing, like data accuracy, technical integrity, inventory management, and payment reliability.
My explanations revolve around broadening SEO’s scope—it’s about ensuring the business is machines-readable, trustworthy, and consistent.
SEO becomes vital in helping leaders identify system or data gaps that could hinder the brand’s selection, highlighting its connection to both risk management and operational resilience.
In most e-commerce brands, agentic systems affect the top of the funnel first. Discovery shifts towards more personalized, conversational interactions.
Instead of brief search phrases, users convey needs, constraints, and preferences, which the agent then transforms into actions.
This decreases the significance of owning category head terms. If an agent has comprehensive user data, it acts like a knowledgeable repeat customer.
This presents a new reporting challenge. Not all SEO work will appear as direct demand creation, yet it still impacts outcomes. Leaders need to anticipate this shift.
Rethink Consideration
The consideration phase evolves too. Traditionally, it involves hosting reviews, comparisons, and reassurances.
With agentic intervention, consideration morphs into a filtering process, retaining only the options that align with user preferences.
This necessitates a quality over quantity strategy in content, emphasizing structural trust signals and consistent, verifiable information.
Brands might be selected without user awareness. While this could boost conversions, it also poses a risk to brand recognition if not addressed elsewhere.
Measurement often concerns executives, and agentic AI complicates this. With more processes happening inside AI, fewer interactions leave traceable or clear data.
I address this early by stressing that while this isn’t a failure of optimization, it merely highlights the analytics limits in a complex digital landscape.
The focus should shift to directional indicators and blended performance over precise attribution, acknowledging the new decision-making landscape.
Advocate Proactive, Low-risk Responses
The crux of leadership dialogue is next steps. Fortunately, most appropriate responses to agentic AI carry low risk.
Enhancing product information, eliminating inconsistencies, strengthening reliability signals, and addressing technical vulnerabilities benefit the business now and pave the way for the future.
Building brand trust outside search also plays a critical role. Trusted brands are more likely to be selected by agents performing comparisons.
This strategy reassures leaders that success doesn’t require radical change but calls for focused improvement.
Agentic AI: Focus Shifts, Fundamentals Persist
For us SEO leaders, agentic AI modifies our focus. Instead of solely optimizing for visibility, we aim to protect eligibility, reduce ambiguities, and illustrate influence.
This demands confidence and clear articulation, challenging hype with grounded perspectives. Agentic AI renders SEO more strategic and no less crucial.
Agentic AI isn’t an imminent threat or foolproof advantage. It’s a transformation in decision-making approaches.
For e-commerce brands, the winners are those who stay composed, communicate effectively, and transition their SEO approach from driving clicks to securing selections.
This transition forms the backbone of the current SEO leadership discussions.
I recently came across an intriguing study by SALT.agency, focused on Google’s AI Mode and its citation practices. Contrary to popular belief, this analysis shows that AI Mode doesn’t have a preference for content placed “above the fold.”
After sifting through over 2,300 URLs cited by AI Mode, researchers discovered no link between a text’s vertical position on a page and its likelihood of being cited by Google.
Pixel depth is irrelevant. The study revealed that AI Mode pulls text from all over a page, even from content located thousands of pixels down.
Page layout vs. content visibility. While different layouts like large hero images or narrative formats might push text deeper down the page, this doesn’t impact whether it gets cited.
Subheadings make a difference. One key pattern identified was AI Mode’s tendency to highlight a subheading and the subsequent sentence. This suggests Google’s heading structures are crucial for content navigation.
Google’s approach. The assumption is that AI Mode employs fragment indexing technology, breaking pages into sections and pulling the most relevant fragment, irrespective of its position.
Dan Taylor, a partner at SALT.agency, confirms that there’s no secret formula for appearing in AI Mode citations. The focus should always be on crafting well-structured, authoritative content that meets customer needs.
Our takeaway. This study challenges the notion that specific AI-focused templates or rigid structures enhance content visibility in AI Mode. The real work lies in creating meaningful, structured content.
Research background. SALT scrutinized 2,318 URLs in AI Mode responses. The vertical pixel position of each cited fragment was meticulously recorded using a Chrome bookmarklet and a 1920×1080 viewport.
I realized relying solely on GA4 to assess the impact of AI SEO is like using a broken compass. While GA4 is a great starting point, it doesn’t paint the whole picture.
It’s crucial to look beyond Google’s tools to truly understand how audiences find and choose brands. SEO isn’t just about visits; it’s a journey shaped by algorithms and AI long before visits occur.
Focusing only on measurable visits hides parts of this journey, leaving potential customers adrift. Understanding user intent through share of voice and mapping brand visibility with AI analytics is key.
I’ve learned that measuring AI visits with GA4 begins with tracking sessions from various AI sources. Creating a custom exploration to track these is an important first step.
Despite its ease, GA4 struggles to fully capture AI’s impact. Many AI outputs can’t be distinctly tracked, making it crucial to explore other data sources to get a complete picture of brand impact.
Both Google Search Console and Bing Webmaster Tools don’t separate AI queries effectively, often mixing AI metrics with standard web traffic, making it challenging to gauge AI’s real impact.
I’ve found utilizing regex in GSC to identify conversational queries useful, but as query diversity grows, distinguishing synthetic from human becomes harder.
Exploring AI agent analytics through log files has been insightful. AI agents using text-based browsers evade traditional analytics, requiring SEOs to delve into bot logs for agent patterns without real human traffic miss them.
Following AI agent request paths, especially to conversion pages, reveals broken journeys and insights into improving user paths.
Reassessing traditional SEO reporting frameworks is essential for adapting to AI’s transformational role in search discovery.
We need tools that track in-chat brand mentions and citations beyond standard website links. AI search analytics must evolve, reflecting SEO’s expansion towards measuring meaningful marketing KPIs and increasing market share.
As an SEO, my goal is no longer optimizing just a website. It’s about building a robust digital brand—one that is visible and trusted across all organic surfaces.
Analyzing nearly two million LLM sessions across nine industries throughout 2025 was a fascinating journey for me. I began with the assumption that ChatGPT would dominate and that AI usage patterns would be relatively uniform with minimal impact.
The findings, however, were surprising.
While ChatGPT does indeed control 84.1% of the trackable AI discovery traffic, it’s primarily serving as a broad-market tool. This discovery significantly impacts strategic approaches.
In today’s landscape, relying solely on a single discovery strategy is not viable. A multi-platform approach that aligns with how and where users find productivity is essential.
Brands must now discern which platforms are empowering productivity rather than merely supporting initial discovery phases.
Various LLMs are excelling in different sectors, often with stark differences. The key takeaway for 2026 is more complex than simply focusing on ChatGPT.
Here’s what I’ve discovered from the data.
The Growth Rate Divergence: ChatGPT vs. Competitors
Throughout 2025, major LLM platforms exhibited significant growth discrepancies:
ChatGPT: 3x growth
Copilot: 25x growth
Claude: 13x growth
Perplexity: 1x growth
Gemini: 1x growth
Although ChatGPT grew, Copilot and Claude experienced much more rapid growth. Platforms like Perplexity and Gemini remained steady, reinforcing specific workflows.
These numbers highlight strategic priorities:
Satya Nadella celebrated Copilot reaching 100 million monthly users.
Dario Amodei revealed that Anthropic’s revenue grew from $100 million to $8–10 billion in under two years.
Aravind Srinivas noted significant interest in Perplexity Finance.
The focus on growth is crucial because it signals true user value:
Copilot excels in the Microsoft ecosystem.
Claude appeals to developers.
Perplexity thrives among finance professionals.
Different LLMs are thriving in various industries at markedly different rates.
Pattern 1: Copilot’s Striking Growth
Copilot’s remarkable 25x growth is indicative of its premier position in B2B environments reliant on Microsoft tools.
SaaS
ChatGPT: 2x growth
Copilot: 21x growth
The rapid adoption mirrors modern SaaS practices, embedding LLMs directly into workflows.
Education
ChatGPT: 6x growth
Copilot: 27x growth
Copilot benefits from educational settings fostering knowledge sharing and synthesis.
Finance
ChatGPT: 4.2x growth
Copilot: 23x growth
Finance aligns with Copilot due to automation needs and context dependency.
Copilot’s growth is most pronounced in industries where professionals are deeply integrated with Microsoft tools.
Instruments like Excel transform into data interpretation powerhouses with Copilot, eliminating the need for external searches.
Implications
For work-centric audiences like SaaS, finance, and education specialists, AI discovery is shifting into LLMs embedded in workflows.
Pattern 2: Perplexity Shines in Finance
While Perplexity has flat growth overall, it stands strong in finance with a 24% market share, unlike in other sectors where it has diminished.
SaaS: down to 7.3%
E-commerce: down to 3.4%
Education: down to 5.2%
Publishers: down to 3.6%
Finance demands accuracy; thus, traceable sources make Perplexity vital in this sector.
Partnering with Benzinga, FactSet, and others, Perplexity offers in-depth data vital for financial decisions.
Trust and verifiability are crucial in finance, and that’s where Perplexity excels.
Implications
In finance, selection of platforms that integrate with licensed data and credible sources is critical. Success hinges on being part of these authoritative ecosystems.
Pattern 3: Claude’s Dominance in Analysis
With just a 0.6% share, Claude might appear to be an underdog, but it thrives in specialist sectors like publishing and finance.
Publishers: 49x growth
Education: 25x growth
Finance: 38x growth
SaaS: 10.3x growth
Claude’s strength lies in standalone, strategic thinking rather than integrated tools like Copilot.
Publishing professionals and financial analysts use Claude for its substantial context window, enabling complex and strategic queries.
Implications
Target audiences that require in-depth analysis should focus on creating structured and detailed content. Claude’s user base is smaller but highly influential.
Pattern 4: Challenges in Tracking Gemini
The data concerning Gemini is puzzling, showing both growth and declines. This could be attributed to issues with attribution rather than an actual decline in users.
Education: −67% tracked traffic
SaaS: +1.4x growth
Finance: +1.3x growth
E-commerce: +2.7x growth
Gemini’s interaction model keeps users within its ecosystem, making measurement challenging.
The reality is that usage might still be robust, but the tracking systems need to catch up with user behaviors.
Implications
As AI-assisted conversions increasingly occur, traditional last-click attribution models need reconsideration.
Monitor brand search performance and invest in broader visibility strategies.
Strategizing Your LLM Approach
AI discovery is diversifying rather than converging. Tailoring strategies based on your audience’s preferences and behaviors is crucial.
Enterprise Audiences: Focus on Copilot integration for SaaS and B2B environments.
High-Stakes Decisions: Consider Perplexity’s reliability in providing traceable data.
When I first stumbled upon the concept of query fan-out, I realized how misunderstood it often is in the world of AEO and SEO. It’s fascinating how AI searches can take a single prompt and transform it into numerous sub-queries, expanding the scope of search in unimaginable ways.
Understanding this process opened my eyes to the hidden potential these sub-queries hold. By leveraging the data generated from them, I discovered new strategies to enhance SEO effectiveness, making my digital marketing efforts more robust.
I’ve discovered that refining our approach with both robots.txt and LLMs.txt can significantly enhance our visibility on platforms like ChatGPT, Gemini, and various answer engines. It’s fascinating how these simple files can make such a big difference in crawlability for AI bots.
Join me as I delve into the best practices for optimizing these important files. By making small yet impactful changes, we can ensure our content gets the attention it deserves in the evolving digital landscape.
I recently came across an intriguing study about AI recommendation lists that caught my attention. It revealed that AI systems like ChatGPT, Claude, and Google’s AI don’t often repeat the same recommendations when asked for brands or products. This means if I ask them the same question multiple times, I’ll likely get different lists each time.
This finding came from Rand Fishkin of SparkToro and Patrick O’Donnell of Gumshoe.ai. They investigated how consistent generative AI recommendations are, and their results were quite fascinating.
What They Tested. Over 600 volunteers used 12 identical prompts on ChatGPT, Claude, and Google’s AI nearly 3,000 times. What they found was quite revealing.
Each AI response was turned into an ordered list of brands or products, and the overlaps, order, and repetitions were compared to see how often the same answers appeared.
The short answer: almost never. Achieving identical lists twice was incredibly rare, with odds of under 1 in 100, and getting the same list in the same order was even less likely at 1 in 1,000.
Even the length of the lists varied. Some responses listed only two or three options, while others had more than ten. If I’m dissatisfied with the result, simply asking again might yield a better outcome.
Why This Matters. We often hear about personalization in AI answers, but this study is the first to provide real data to support that claim, showing a clear departure from traditional SEO.
Design and Randomness. This variability isn’t a flaw — it’s intentional. These systems are probability engines designed to create diverse outcomes, not stable ordered results like Google’s blue links.
One Consistent Metric. Despite fluctuating rankings, one metric that proved more stable than expected was visibility percentage. Some brands repeatedly appeared in a majority of responses.
Consistent presence in these lists carries more weight than exact ranking, especially across multiple runs and intent changes.
Context Size Counts. The consistency of AI answers improves in smaller, niche markets compared to larger categories, where results scatter significantly.
Real-World Prompts. Testing with actual human prompts showed varied results — as people phrased their queries differently, semantic similarity was low.
Yet, AI still returned similar brands for the same intent, proving that AI captures the underlying purpose behind the queries.
The Power of Intent. Even with hundreds of unique prompts for headphone recommendations, prominent brands like Bose, Sony, and Apple surfaced consistently.
When I change the purpose — say, to gaming or noise-canceling — the brand results shift accordingly, indicating that AI comprehends intent despite varied prompts.
What Doesn’t Help. Tracking exact positions in AI answers is unreliable because these rankings are too unstable to mean anything.
What Could Work. A more effective approach might be to track how frequently my brand appears over many prompts, even if it seems complex and imperfect.
Unanswered Questions. There are still gaps to explore, like determining how many attempts are needed for reliable visibility stats or whether API-based results align with real user behavior.
Conclusion. AI recommendation lists are inherently variable, but with large-scale, careful visibility measurement, I can derive actionable insights. Just don’t mistake this for traditional ranking metrics.
For more details, you can read the full report here.
