BrightonSEO talk roundup April 2026

This April, some of the Varn team headed to Brighton for the world’s biggest search marketing conference, brightonSEO.

We’ve rounded up summaries of the most insightful sessions, so you can catch all the key takeaways from the event.

Kyle Rushton McGregor provided a practical guide to identifying and resolving tracking failures. The session emphasised the importance of proactive monitoring and discussed a range of debugging techniques to catch issues before they impact critical business reporting.

Key takeaways:

• Proactive monitoring is key! Use monitoring techniques (such as setting up email notifications) to be notified of tracking breaks and anomalies immediately, rather than waiting for data gaps to emerge later down the line. 
• Maintain a rigorous record of changes to the site and to GTM by using annotations. Tools like the GTM Fixer extension can automatically send annotations to GA4 to provide context for data fluctuations due to tracking changes.

James Yorke discussed how SEO hasn’t stopped working, but the model for measuring it has. In his talk, James advocated for using a three-tiered model for assessing SEO impact; measuring brand presence and user preference alongside performance metrics to demonstrate the full value of SEO.

Key Takeaways:  

• Traffic doesn’t show the whole story anymore, we can interpret data better by using a combination of metrics. 
• Visibility metrics such as rankings, AI citations, SERP features, combined with behaviour indicators such as repeat visits, direct return visitors and brand searches should be added as layers to reporting to help to complete the picture of performance.

Bogdan Babiak presented a large-scale analysis of 100,000 websites to debunk industry myths regarding AI visibility. He demonstrated that traditional SEO remains the strongest signal for AI visibility.

Key Takeaways:  

• The research identified a high correlation (r = 0.720) between good SEO performance and AI visibility, and that SEO-AI correlation has grown over the past year. 
• From Babiak’s research, he found 60% of traffic originating from AI search hits a site’s homepage, highlighting that AI citations act more as a branded experience. 
• The talk emphasised managing “review presence” across Google Maps, social media, and third-party review platforms, as these serve as critical proof points for AI. 
• To counter the current “measurement gap,” the session recommended building internal feedback loops, such as self-reported attribution (“How did you find us?”) and tracking revenue by acquisition cohort.

Tom’s talk focussed on prioritising ranking for terms that actually have organic presence – there’s much less value ranking Position 1 if the searcher has to scroll all the way down past ads and AI features to see you.

Key Takeaways:  

• Look at pixel position in combination with ranking position.
• Prioritise keywords where there’s less scroll (or no scroll!) required to be “seen” in Position 1.
• Sometimes, a keyword in Position 2 has more traffic potential if it’s in a higher pixel position in the SERP.

Janaina‘s talk focused on measuring AI performance and moving away from the more traditional performance metrics like clicks and sessions.

Key Takeaways:  

• Organic clicks are going to be down due to AI search.
• Focus on reporting on revenue (or leads) instead. That’s the best indicator of what’s actually happening.
• Pages with slower speed have fewer citations.
• Test AI model data and AI search data separately, you’ll see different results.

Jake’s talk focused on client relationships. In the first 90 days with a new client, agencies should focus on proper onboarding and completing ‘quick wins’. Account teams should seek regular feedback and adapt communication styles for different types of stakeholders. Stop trying to be the best; be the agency they can’t imagine losing.

Key Takeaways:  

• Client relationships determine what gets implemented. Clients are more likely to implement recommendations of agencies they trust, so it’s important to build those relationships early on.
• Recommendations only move as fast as the internal team, make sure you have a client champion to help push things through with internal stakeholders (senior leadership, reviewers, dev).
• SEO results take longer than clients expect, so goals and KPIs should be set in the kick off call to make sure expectations are aligned, and it’s even more important to build those strong relationships early on.
• Use bi-weekly calls, clear agendas, and in-person meetings to focus on metrics the client cares about.

Liv Day addressed the reality of declining organic traffic, providing a framework for how agencies can discuss these trends with clients proactively before being asked.

Key Takeaways:  

• Zero-click search is the new normal.
• Talk about drops in traffic before you’re asked about it by your clients.
• If blog traffic is down but revenue is stable, it likely means users are using LLMs for informational intent while your commercial traffic remains relevant.
• Pivot your reporting. If traffic is lost but revenue is stable, change the focus of SEO success from traffic to transactions.
• If both traffic and rankings are down, investigate competitor activity and focus on ranking for more commercial terms.

Anu Ramani challenged the industry to stop focusing on the speed of content creation and instead build AI systems that move meaningful business metrics. She introduced the MAP (Message, Architecture, Process) system for creating high-quality AI-assisted content.

Key Takeaways:  

• The MAP System: Focus on Message (differentiation and creativity), Architecture (taxonomy, metadata, intent mapping), and Process (scalable AI-style workflows).
• Use AI for the research phase, understanding what buyers care about and what competitors are missing, to ensure content remains differentiated.
• With 95% of organisations seeing no ROI from AI, the focus must shift from volume to business impact.

Justyna Brownbridge discussed the importance of activating internal experts, noting that 61% more people now use LinkedIn as a search engine than Claude. She advocated for employee advocacy as a “human edge” in an AI-dominated landscape.

Key Takeaways:  

• LinkedIn is the #1 source AI models cite for professional knowledge.
• Activation should start with leaders, then senior leadership, then the wider employee base.
• Don’t ask experts to write content; ask them to talk and then extract the insights (“the gold”) for publication.
• Only 3% of users share content on LinkedIn; those who do regularly can significantly boost brand visibility and trust.

Annika Haataja presented a qualitative “Content Scientist” framework designed to create content that is easily absorbed by both users and AI

Key Takeaways:  

• Content must have a clear Purpose, a Structure that is easy for AI to absorb, clear Value for the user, and established Credibility.
• Authority is earned by fulfilling specific user needs that your organisation is uniquely qualified to address.

Sean Barber shared how Macmillan Cancer Support tackles misinformation in the Your Money or Your Life (YMYL) sector by controlling the narrative with expert, authoritative content.

Key Takeaways:  

• Use expert reviewers and display trust signals, such as the PIF TICK from the Patient Information Forum.
• Create content that maintains its effectiveness even when summarized or shared by AI.
• Maintain a visible correction loop and transparent production processes to ensure reliability.

Daniel Liddle argued that winning in the age of AI search depends on frameworks and human judgment rather than the size of a company’s technology stack.

Key Takeaways:  

• Tools are not a strategy; automation often leads to “sameness”.
• Align audience needs, business value, and AI visibility into one cohesive approach.

Olabinjo Adeniran highlighted the often-overlooked value of using SEO for customer retention and loyalty, noting that acquisition costs five times more than retention.

Key Takeaways:

• If you don’t provide post-purchase content, unverified sources on Reddit or YouTube will fill the gap, potentially leading to a poor customer experience.
• Identify content needs by looking at customer support logs, online forums, and product telemetry.