BrightonSEO 2025 San Diego brought together hundreds of SEO professionals, marketers, and business leaders to confront one pressing question: how is search evolving in an AI-driven world? Over two days at the Manchester Grand Hyatt, industry thought leaders and vendors shared insights that spanned beyond Google rankings, from AI visibility and generative search, to schema, technical SEO, and content diversification.
For practitioners, the conference offered tactical takeaways; for businesses, it revealed where budgets, workflows, and public opinion are shifting. This recap explores not only the standout sessions, but also the vendor debates and networking moments that show why attending SEO conferences remains essential in 2025.
Why SEO Conferences Still Matter:
Conferences remain critical touchpoints for the SEO industry. They compress months of change into concentrated
sessions and provide exposure to real-world experiments you can’t always see in case studies or online forums.
For professionals, the value lies in validating strategies, testing new frameworks, and expanding networks. For businesses, events provide clarity on where the industry is heading, which tools are shaping public opinion, and how peers are responding to generative search and AI disruption.
From the democratization of search beyond Google to the urgency of AI adoption and the continued importance of technical SEO and schema, the discussions highlighted both challenges and opportunities shaping the next phase of our industry.
Here are my takeaways from those conversations, and the steps I believe are necessary to unpack, test, and prove real value and ROI for clients moving forward.
The Democratization of Search: Beyond Google
A recurring theme across sessions was that search is no longer just about Google.
Ashley Lidell emphasized the idea of “search everywhere optimization,” encouraging brands to expand their content beyond Google to platforms where users actively engage. This echoed talks from Wil Reynolds and Ross Hudgens at both BrightonSEO and SEOweek.
Reynolds’ point was particularly striking: brands may appear to be “winning with search, but losing with people.” He argued that while visibility on Google can look positive, those metrics often mask underlying problems. When users don’t feel they’re getting trustworthy information, they turn to other platforms, Reddit threads, TikTok, YouTube reviews, where sentiment carries more weight.
Ricardo Baeza-Yates reinforced this at SEOweek, noting that intent is shifting toward apps and predictive channels. The takeaway: SEO cannot remain siloed. To truly meet users where they are, strategies must span search, social, PR, and paid.
Content Diversification and Multi-Platform Ecosystems
Several sessions reinforced the importance of diversifying content across platforms.
Josh Blyskal presented research showing that brand affinity drives AI visibility. Mentions in communities like Reddit and YouTube matter more than raw website traffic in shaping LLM responses. Ross Simmonds, whose work has focused on the value of Reddit for brands, echoed this during his talk in SEOWeek, with a call to “create once, distribute forever;” while Frank Olivo highlighted the critical role of audience retention in YouTube optimization, a metric used to analyze whether users are sticking around to finish the video.
Building a Smarter Content Ecosystem
I found the conference discussions reaffirmed the growing need for content diversification, but what still feels underdeveloped is the framework for building true content ecosystems. SEO can no longer operate in silos — content needs to live across multiple channels and formats, each mapped to a specific user intent.
My approach focuses on cognitive response - understanding how users behave and feel when they search. I’ve found the best way to do this is through question-based queries, particularly People Also Ask (PAA) results.
- Tools like AlsoAsked.com and SEMrush are useful for uncovering question clusters that reveal emotional triggers and intent.
- Unlike traditional keyword lists, question data reflects user perspective, helping identify what motivates a search and what reassurance users seek.
Intent is perspective, and understanding language at that level is how large language models (LLMs) predict user needs.
From a technical standpoint, PAAs and rich results (local packs, featured snippets) show how Google anticipates follow-up questions to shorten what Dan Taylor describes as Time to Result (TTR), the time it takes for users to feel satisfied with an answer.
I asked Dan if TTR connects to query fan-out, where search systems test multiple pathways to find the best result. He explained:
“Query fan-outs represent a form of query stacking, refining searches through multiple attempts. TTR, however, is more about how content is structured and how search engines interpret engagement signals like long clicks versus short clicks.”
He illustrated this with a sharp example:
“In the Arsenal vs. Molde example, the page’s main purpose wasn’t to answer the query, it was to get the user to view as many ads as possible before reaching the answer.”
That distinction reinforces how I think about content ecosystems. Search engines influence how Retrieval-Augmented Generation (RAG) systems in LLMs stay current. Every PAA or featured snippet can act as a grounding signal, helping models retrieve and validate new information.
To build an effective content ecosystem, content shouldn’t just answer intent, it should anticipate it. Whether through text, video, Ads, or short-form formats, the goal is to understand where and how users seek relevance, then design content that meets them with both purpose and context.
Technical SEO as the Foundation of LLM Optimization
One of the strongest themes at BrightonSEO was the shift from simply talking about AI to actually applying it. The mood across keynotes and conversations carried a clear sense of urgency: our industry can’t afford to stay stuck in theory while workflows, budgets, and user behaviors are already being reshaped.
Britney Muller’s keynote stood out. Her phrase, “courage to implement imperfectly”, reframed AI not as a shiny distraction but as a set of tools that need to be tested now, even if the results aren’t perfect. She highlighted practical examples: Phantom Buster for LinkedIn prospecting, WordCrafter for content workflows, Ollama LLMs for local experimentation, and NotebookLM or Google Labs for research and ideation. Her point wasn’t about hype cycles; it was about community, experimentation, and the bravery to build while the paint is still drying.
This pragmatic tone was echoed by Raycheal M. Proctor, Tom Mansell, and Alex Halliday of AirOps, who showed how applied AI can streamline SEO processes rather than replace them. The discussions at SEOweek went further, stressing that training AI agents with SEO expertise is the only way to move beyond generic outputs and into specialized, applied practice.
It’s also important to acknowledge the noise. Every week brings a new acronym, with GEO (Generative Experience Optimization) being most common. While GEO reframes SEO principles for an AI context, its underlying mechanics remain the same, accessibility, structure, and relevance. Whether we call it GEO or SEO, the fundamentals still matter.
At Chartis, the team focuses are fundamental questions behind user journeys and behaviors:
- How are users finding us?
- What happens once they land on our site?
- Does our brand message resonate, evoking a feeling?
These are the questions that have always mattered, regardless of whether the channel is Google, YouTube, or an LLM.
The challenge, and opportunity, is that so much about LLMs remains uncertain. Unlike Google Search, where GSC offers user data at query level, LLM responses are inherently personalized and unstable. Change the phrasing of a
prompt, and the output shifts. We can simulate prompts, test with synthetic ICPs, or analyze citations, but none of this perfectly mirrors real-world user behavior. That uncertainty is precisely why experimentation and practical use matter most.
Muller’s talk reminded us that LLMs aren’t new to search; Google has been layering them into its systems for years, from Hummingbird to Penguin, BERT, and MUM. Today’s reality is that LLMs are expensive to train, and RAG workflows are essential to keep them relevant and grounded.
For practitioners, this reinforces a key truth: SEO isn’t dead, but reactive SEO is. The principles of accessibility, structure, and relevance that we’ve honed for decades are exactly what make LLMs usable. What needs to end is the reactive posture that waits for search changes to dictate strategy. The future belongs to those who adapt proactively, integrating AI as a working, testable component of their SEO practice.
Schema, Entities, and Semantic Relevance
Schema’s role in modern SEO sparked lively discussion. Some argue it’s becoming less necessary as Google improves at interpreting semantics on its own. I see it differently. Schema continues to add value by clarifying entity relationships, exposing structural gaps, and strengthening contextual understanding. Even if LLMs embed schema into vectorized representations (as Andrej Karpathy has noted), schema benefits remain critical as retrieval-augmented systems evolve.
For AI Visibility, I Focus On How Content Communicates Meaning
- Define your semantic keywords clearly within the text - In my interview with the Simple and Smart SEO Podcast, I mentioned the idea behind semantic triples, semantic triples are machine understanding of language using subject - predicate - object to define relationships between concepts. Similar to structure data, these concepts are not a silver bullet, but can be useful when audited against cognitive response to complex and frustrating subject matters.
- Use structured data not as a silver bullet, but as a tool to expand or analyze how clearly your entities are represented on the page.
- Improve internal linking - one of the most critical yet overlooked areas. Many site structures lack semantic clarity. Consider how your landing pages are interconnected to create a logical path of discovery and relationship. Ensure that each blog topic is contextually relevant to your website’s business or product/service offering, providing both users and bots the information needed to reinforce action.
- Remember: Semantic optimization goes beyond schema markup. It’s about giving meaning and context to the relationships between different concepts.
Testing and Validation: Applying the Scientific Method
One area I wish had received more attention at BrightonSEO was testing and validation. The conference featured strong tactical discussions around data, experimentation, and strategic frameworks, all valuable, but what felt missing were real studies or measurable experiments that show how these ideas perform in practice. The theory was strong; the empirical follow-through was limited.
As SEO evolves alongside AI and generative search, we need more evidence-based insights that help us understand whether SEO and GEO are truly distinct disciplines or simply different expressions of the same principles. Much of what was shared still leaned on long-standing SEO fundamentals, valuable, yes, but familiar.
In this new landscape, reliable testing frameworks and data integrity are more critical than ever. Without credible validation, our understanding of what works, and why, risks remaining theoretical rather than actionable.
At Chartis, we take the same approach. Our team prioritizes testing over noise, focusing on the data that truly matters. SEO works in partnership with Technical, CRO, UX, and Analytics teams that are equipped to run real experiments and uncover insights grounded in evidence.
As my colleague Mathieu Sussman put it:
“There’s a lot of data out there, but none of it really means anything.”
Collecting the right data, and ensuring it’s usable, is what turns testing into actionable insights that scale.
Looking Forward: Emerging Trends Shaping SEO
- Diversified ecosystems — content that lives across channels and aligns with user intent, emotion, and format.
- Semantic precision — using schema, internal linking, and structured data not as a checklist but as tools to clarify meaning and context for both users and machines.
- Scientific rigor — emphasizing testing, validation, and data integrity to separate signal from noise and ensure SEO remains evidence-based in an era of uncertainty.
- Human-centered analysis — focusing on cognitive response, sentiment, and engagement as indicators of trust and relevance beyond rankings.
- Integration with AI workflows — recognizing how LLMs, RAG systems, and generative experiences reshape discoverability, demanding proactive adaptation rather than reaction.
Reflections on the Conference Circuit and the future on Content Strategy:
Attending SEOweek (NYC), WTSFest (Philadelphia), and BrightonSEO (San Diego) offered not only tactical updates but also perspective on the conference circuit itself. The camaraderie across these events was unmistakable, yet many sessions either reinforced what we already know or introduced emerging ideas and technologies that haven’t yet found their footing in SEO. Either way, I left wanting more — and that’s a good thing.
My perspective on SEO remains consistent:
- Leverage user intent and sentiment. Use NLP systems such as LLMs or Google’s NLP API (Sentiment Analysis) to measure and interpret cognitive responses when relevant. As user behavior evolves, aligning creation and optimization with your brand message and audience mindset is essential. Delivering a product or service that people genuinely want to discuss remains central to how these systems interpret and model language.
- Understand how these systems work. Ensure your content isn’t just present — it’s semantically optimized for extraction and meaning. Concepts like Time to Result (TTR) and Query Fan-Out can help expand or structure content for multiple purposes and contexts.
- Test your ideas. It’s not enough to say “It’s SEO” or “SEO is dead.” Validate SEO concepts against LLMs to understand what truly drives discoverability and where your time delivers the most value for clients.
Across every event and conversation, one truth stands out: SEO is no longer a channel, it’s an evolving system of meaning, experimentation, and intent. The path forward isn’t about chasing algorithms or new acronyms, but about testing, validating, and connecting insights that drive measurable impact. As search continues to blend with AI, the SEOs who will lead the next chapter are those who understand how to balance structure with creativity, data with intuition, and most importantly, how to translate complexity into clarity for both users and machines.
Are you preparing for AI Search?
At Chartis, we help clients build discovery strategies for an evolving search landscape shaped by AI. If you’re looking to improve visibility as search moves beyond traditional rankings, we can help.