Agenda 2026

Join Johan von Hülsen for a relaxed 5K before the conference kicks off. No pace pressure, no performance targets. Just Berlin in the early morning, a clear head, and your first connections made before the conference has even started. We will meet in the hotel lobby at 7:00 am. All paces genuinely welcome.

Search authority is undergoing its biggest transformation since the early days of PageRank. As AI-driven systems increasingly evaluate brands, experts and claims across multiple sources, trust is moving beyond links and domains into a broader ecosystem of signals: expertise, brand reputation, social validation, third-party citations and proven usefulness. This keynote explores how brands must evolve from chasing rankings to building trust at scale.

As Generative Information Retrieval reshapes the search landscape, the SEO industry is facing a steep learning curve and a wave of new misconceptions and even more industry myths. In this session, Dawn will dismantle emerging SEO myths and clarify complex new concepts. Discover the true origins of these misunderstandings and learn exactly what generative search actually means for the future of your SEO strategy.  The session delves into the ‘AI SEO misinformation problem’,  examining why the field is prone to subjective interpretations and how the emergence of Large Language Models (LLMs) has exacerbated longstanding debates. By demystifying how search engines actually rank and surface content versus how they are engineered internally, this talk seeks to separate technical reality from the ‘echo chamber’ of trending terminology.

The measurement problem for senior performance marketers isn’t a lack of data — it’s too many sources, too many conflicts, and too little clarity on what’s actually driving growth.

In this session, we’ll discuss the interplay of causal and correlative data and how to orchestrate insights across MMM, incrementality experiments, and attribution solutions for more profitable media optimisation. We will dive into building a resilient measurement stack, tapping AI tools for better analysis, and how to frame measurement findings in terms that land with a CFO to actually influence long-term budget decisions.

You’ll hear Ginny’s perspective on the current state of Meridian MMM, incrementality experiments, Google Analytics, and other measurement solutions, and what a rigorous, joined-up approach looks like in practice.

ChatGPT, Perplexity, Gemini, Claude, Copilot: SEO for LLMs may seem interchangeable – but these platforms are built on very different foundations. Creators, investors, media partnerships, retrieval systems, training data and knowledge cut-off dates shape how each model selects sources, frames answers, and surfaces brands.

Viola will break down what differentiates the leading LLMs today and why those differences matter for search visibility. You will learn how to think in LLM ecosystems, how to adapt optimization approaches across models, and how to stay ahead as the balance of power shifts over time. This is about making informed decisions in a landscape where AI systems, not search results, increasingly control exposure.

ChatGPT ads are moving from theory into real-world execution. Enterprise advertisers are already testing the channel with meaningful budgets and clear performance expectations.

In this session, Sam shares first-hand insights from clients actively running ChatGPT campaigns. What formats are emerging, how targeting behaves, where performance diverges from search and social, and what breaks when familiar playbooks are applied. We’ll look at setup realities, measurement limitations, and how teams are adapting to a channel that doesn’t behave like traditional paid media.

If you’re responsible for serious media budgets, ChatGPT is not just another placement to add to your mix. The mechanics, signals, and user behavior are fundamentally different. This session offers an early, experience-based view of what that means in practice.

Indexation used to be simple. Five years ago, it started becoming difficult. Today, it is expensive, and therefore limited. At the same time, the amount of content continues to grow. AI systems require more and more longtail content to generate answers, yet they do not maintain their own index. Instead, they rely on the indexes of Google and Bing. That raises a critical question: what does it actually take to be indexed and to stay indexed, in this environment?

Is crawl budget really the core problem? What truly leads to indexation? What causes de-indexation? And most importantly, what concrete actions can you take to improve your chances of being indexed and remaining visible?

In this session Johan, looks at the mechanics behind inclusion and exclusion, challenges common assumptions, and focuses on practical steps you can implement to improve indexation in a world where content is abundant, AI depends on existing search indexes, and visibility is no longer guaranteed.

AI Max for Search introduces powerful automation, but many even advanced advertisers miss many opportunities to make this campaign type work to optimize profits? Join scripting wizard Nils Rooijmans, Google Ads Performance Architect, as he shares the latest AI Max strategies. You’ll learn how to use scripts to reduce wasted ad spend, control your brand, monitor performance, and adjust settings when needed

After this session, you’ll be able to understand the algorithms behind AI Max and what role this campaign type plays in AI Overviews, translate your business objectives into a setup for your ‘AI Max for Search’ campaigns and use a set of provided tactics and scripts to optimize your campaign setup to increase _your_ profit.

The people sitting across from you have the same job, the same pressures, and probably some of the same unsolved problems. These roundtables are consistently the highest-rated part of SMX Advanced. Not because of who’s presenting, but because of who’s sitting across from you. Senior practitioners, in-house leads, agency heads – people managing real accounts, real teams, real budgets. Off the record. No pitches, no positioning. Just people willing to say what they actually think.

Each table works from a specific, pre-defined problem. There’s a facilitator, a shared document, and three concrete takeaways per group that feed back into the room. Come with something real. A method that worked. A decision you’re still not sure about. A problem you haven’t solved yet.

These are your people. The conversation starts the moment you sit down.

Tables:
1. GEO measurement: what are you actually reporting, and do you trust it?
Everyone has cobbled together something. Prompt sets, citation tracking, share-of-voice proxies. This table is for comparing notes honestly — what your setup looks like, where it breaks, and what you’ve stopped bothering to measure because it told you nothing useful.
2. Brand representation in LLMs: what do you do when the sources getting cited aren’t you?
You’ve checked. ChatGPT mentions a competitor. Perplexity cites a review site with outdated information. Your own content doesn’t appear. This table is about the practical response — how teams are identifying where the gap is, what levers they’re pulling, and whether any of it is measurably working.
3. The long-tail content question: keep, consolidate, or let it go?
AI answers a huge chunk of informational queries now. A lot of content that used to pull traffic no longer does — and won’t again. This table is for the decision-making process: how are you auditing what’s worth keeping, what criteria you’re using, and how you’re making that argument internally without it turning into a six-month project.
4. Who owns GEO — and how is that working out?
In some teams it sits with SEO. In others it’s been handed to content, brand, or a new “AI visibility” function that didn’t exist 18 months ago. This table isn’t about org design theory — it’s about what’s actually happening: the workflow gaps, the ownership disputes, the things falling through the cracks because nobody’s sure whose job it is.
5. Open table: AI in search — what’s your biggest unresolved question right now?
No fixed agenda. Bring whatever is keeping you up at night. The only rule: it has to be a real question, not a talking point.

The people sitting across from you have the same job, the same pressures, and probably some of the same unsolved problems. These roundtables are consistently the highest-rated part of SMX Advanced. Not because of who’s presenting, but because of who’s sitting across from you. Senior practitioners, in-house leads, agency heads – people managing real accounts, real teams, real budgets. Off the record. No pitches, no positioning. Just people willing to say what they actually think.

Each table works from a specific, pre-defined problem. There’s a facilitator, a shared document, and three concrete takeaways per group that feed back into the room. Come with something real. A method that worked. A decision you’re still not sure about. A problem you haven’t solved yet.

These are your people. The conversation starts the moment you sit down.

Tables

1. Automation boundaries: where are you holding the line, and why?
Every account has a different answer. Some teams are locking brand, negatives, and budgets and releasing everything else. Others are doing the opposite. This table is about the actual decisions — what you’re controlling, what you’ve handed over, and what made you draw the line where you did.
2. Proving incrementality without blowing up the account: what’s your approach?
Everyone agrees it matters. Far fewer have a method they’d stand behind. This table is for sharing what you’ve tried — holdout tests, geo experiments, MMM — what the results actually showed, and how you explained them to someone who just wanted to see ROAS go up.
3. AI-generated creative at scale: where have the guardrails failed you?
Not a theoretical question. This table is for the incidents — the ad that went live and shouldn’t have, the brand guideline that didn’t survive contact with automation, the approval process that couldn’t keep up. What broke, what you fixed, and what you’re still not confident about.
4. Signal loss and consent: what’s your current tracking setup, and are you comfortable with it?
Server-side, consent mode, modelled conversions — most teams have patched something together under real GDPR pressure. This table is for comparing notes on what’s actually holding up, what’s quietly broken, and what you’d do differently if you were starting from scratch today.
5. Open table: paid search in 2026 — what’s the question you can’t get a straight answer to?
No fixed agenda. Bring the thing you’ve been arguing about internally, or the thing nobody in your organisation wants to address yet. Peer conversation, no vendor spin.

Reddit has become one of the most influential platforms for organic visibility in Google and LLM-driven search — and one of the most unforgiving environments for corporate brands.
In this session, Benjamin shares how he and his team built a sustainable Reddit SEO strategy under real-world corporate constraints. From taken usernames and moderator roadblocks to legal review cycles and internal governance, he explains what it actually takes to operate on Reddit without triggering bans, backlash, or long-term brand damage. You will also learn what it means to engage on Reddit in a way that is perceived as genuinely helpful rather than corporate, how to navigate community and moderation dynamics, and how to build a community setup that earns visibility and trust, and ranks in both Google and LLM results.

Generative AI has given brands a superpower: the ability to create personalized creative at infinite scale. But superpowers create super-problems. How do you test a million variations? How do you know why Ad #347 worked for Gen Z while Ad #892 worked for Boomers? How do you avoid destroying your brand with AI-generated garbage? This talk introduces creative intelligence as the emerging discipline that transforms generative AI from a production tool into a strategic weapon, and explains why 2026 will be the year every CMO has to master both.

Search no longer runs on a single track. Organic, paid, and AI-driven discovery now overlap, but most teams still plan and measure them separately. In this session, Anu Adegbola and Jono Alderson speak with brands about how they are dealing with that reality. You’ll get a direct look at how teams are restructuring strategy, budgets, and workflows to reflect how search actually works today. The conversation focuses on practical questions: how to evaluate visibility across AI and traditional search, how to move beyond last-click thinking, and how to align SEO, PPC, and AI efforts into something that compounds instead of conflicts. You’ll leave with concrete ideas you can apply, grounded in how other teams are solving the same challenges right now.

After the first day of sessions wraps up and the networking reception winds down, we’re heading out for dinner.

Here’s how it works: we’ll meet at 7:00 PM in the hotel lobby, then split into SEO and PPC groups and walk to nearby restaurants — tables already reserved. Whether you end up with the paid search crowd or the SEO contingent, the format is the same: good conversations, no agenda, good food.

This is exclusively for registered SMX Advanced attendees. No additional signup needed. Dinner is pay-your-own-way — we handle the logistics, you just show up.

The conference is in full swing, and this is one of the best ways to start Day 2. Join us for another relaxed 5K with Johan von Hülsen. No pace pressure, no performance targets. A clear head, easy conversation, and a chance to make new connections or pick up where yesterday’s left off.

We meet in the hotel lobby at 7:00 am. All paces genuinely welcome.

Your rendering strategy was probably decided by your engineering team, for your users. Sam Torres makes the case that it’s also one of the most consequential SEO decisions you’re not actively managing. That gap is only growing as AI crawlers and agents become a meaningful source of traffic and visibility.

This session goes beyond traditional crawler considerations to address two converging shifts: the growing popularity of LLM-driven search and the coming agentic web. Sam maps each rendering approach to its real-world impact on discoverability, structured data accessibility, and agent compatibility, and offers a practical framework for auditing and future-proofing your site for what comes next.

You’ve been playing the bidding game for years. You know how it used to work. But what once delivered reliable results is now being reshaped by automation, AI, and shifting platform dynamics. Today, you have to challenge every assumption you’ve ever made about bidding.In this session, Jyll reexamines the foundations of Google Ads and rewrites the playbook for advanced bidding. She’ll dispel common bidding myths, question long-held best practices, and show how to rethink control, signals, and optimization in a landscape where automation is no longer optional.

The way we find, filter, and act on information is being fundamentally transformed by AI agents. No longer limited to retrieving links or ranking results, today’s AI systems can understand intent, synthesize vast amounts of data, and take action on behalf of users. In this keynote, Farah draws on her experience building AI-powered search and intelligence products at Feedly to explore how autonomous agents are reshaping search, from passive keyword matching to proactive, context-aware information delivery. She’ll share practical insights on building agents that don’t just find answers but analyze, prioritize, and act, and what this shift means for search professionals, marketers, and organizations looking to stay ahead.

AI Search is often discussed through patents, predictions or opaque model behavior. This session takes a different route: real evidence. Using Chrome DevTools, network logs and script analysis, we will examine how ChatGPT, Copilot and Gemini actually operate behind the scenes. You will see how request patterns, embedded metadata, source handling and internal feature flags reveal concrete mechanisms that shape AI-generated answers. The talk demonstrates a practical methodology for identifying signals, understanding system logic and uncovering hidden functionalities. Attendees will leave with a technical framework they can use to analyze AI search systems independently and extract insights that inform strategy, optimization and competitive research.

AI isn’t just another tool in the paid media stack – it’s changing how we build the stack itself. In this advanced session, Matt goes deep on using AI to create custom tools that genuinely improve how you work.

You’ll see how multi-model workflows – passing specs, plans, and code between Claude and OpenAI – produce sharper, more reliable outputs than any single model alone, and how this approach applies directly to paid media problem-solving.

The session covers how AI is reshaping software development for marketers who aren’t traditional developers. Matt will draw a clear line between two modes of thinking: AI as a productivity tool, and AI as an enhancement tool that combines your strategic ideas with platform data to generate outputs at a scale you just couldn’t produce manually.

Expect live examples, working code, and practical frameworks you can take back and start building with immediately.

Many GEO teams still evaluate AI visibility with a few prompts and a tracking screenshot. That is no longer enough. Across ChatGPT, Gemini, Google AI Mode, Grok, Perplexity and other systems, answers, cited sources and grounding behavior can vary dramatically. In this session, I will share a vendor-neutral framework for classifying prompts by grounding likelihood before generation, measuring answer and citation stability, and distinguishing cited from non-cited source patterns. I will show where GEO tracking breaks in practice, what actually drives retrieval and citation, and how search teams can build a reliable testing and optimization workflow for AI search visibility.

Most platforms tell us “go with the flow” and trust the algorithm – invest Google’s PMax or Meta’s Advantage+. But this blind trust is a recipe for disaster. Platforms push what they think people want to see, wasting your spend on “window-shoppers” while neglecting promising products. The fight for control has moved. It’s no longer about audience buttons or old-school segmentation; it’s about the data you feed the machine. This session provides a playbook for ecommerce advertisers to force the algorithm to flow where it actually provides incremental value for your brand.

Most ICP documents are built once and trusted too long. This session shows what happens when you actually interrogate them with data. Ashley shares how SearchLab rebuilt their ICPs using behavioral intelligence, first-party marketing data, and AI-assisted analysis — and what they found contradicted several existing assumptions. Who consumes content versus who actually influences buying decisions. How target audiences describe their problems in search queries. Where messaging frameworks diverge from real decision-stage concerns. You’ll see how those signals were triangulated, what they pointed to, and how they translated into concrete changes in content strategy, topic prioritization, and search alignment.

At scale, performance data doesn’t just become less reliable, it starts to distort the decisions built on top of it.

A stable CPA can mask volatility that erodes profit. ROAS can improve while contribution margin declines. Attribution models consistently reward what is easiest to measure rather than what actually drives incremental outcomes. In large accounts, these effects don’t cancel out, they compound.

This session assumes you already know the limitations of standard metrics. Instead of revisiting them, it focuses on six specific statistical failure modes that emerge in high-spend Google Ads environments and quietly bias optimisation over time.

Drawing on analysis of $770M+ in ad spend, Igor will walk through how these distortions show up in real datasets, how to quantify their impact, and how to recalibrate your evaluation frameworks accordingly. The session connects variance, distribution effects, and incrementality with the day-to-day decisions that shape account performance, with a focus on separating apparent efficiency from actual economic impact.

You’ll leave with a diagnostic lens for interrogating your own data more rigorously, and a clearer understanding of when performance improvements are real, and when they are artefacts of the way your metrics are constructed.

Google’s shift toward AI-powered ad formats has changed what “good” lead gen looks like. Default settings and old “best practices” aren’t enough anymore! This session covers the advanced playbook for lead gen accounts: where Demand Gen, AI Max and Performance Max actually fit (and where they don’t), and how to use micro conversions to teach the algorithm what a qualified lead looks like. We’ll look at what’s working in practice across accounts large and small, including the operational changes needed to get full value from these newer formats. You’ll leave with specific tactics to stress-test and sharpen the lead gen setups and elevate them to the next level.