A Round-up of All the Great Talks From MozCon London 2025

Last week, we hosted the first MozCon London conference, and it was a blast. Ten incredible speakers delivered actionable talks on AI search, brand visibility, and new SEO metrics to show value.

We’ll turn the talks into full-length blog posts in the coming weeks, but for now, here are the key takeaways.

PS: We’re headed to New York on November 6, 2025. Grab your spot now before tickets run out. If you need help getting buy-in, here’s a free pitch kit to convince your boss to send you to MozCon.

1. The infinite tail: Keyword research for AI – Dr. Pete Meyers

Key takeaways:

• Natural language queries are the future of search 

Search is becoming more conversational and textual. SEOs must adapt by aligning content with question-based patterns.

• All keyword variations don’t need tracking—cluster them by meaning

You have to assume each search in an LLM has a volume of one. It makes more sense to cluster them by meaning and use a keyword research tool like Moz to group related queries, instead of chasing individual keywords.

• Focus content in spaces where AI is less dominant

AI overviews are increasingly answering informational searches, but commercial and transactional queries are still monetized by ads. Shifting content strategy toward mid-funnel and editorial topics may protect traffic and drive conversions in the interim.

2. How to drive more conversions with fewer clicks – Rebecca Jackson

Key takeaways:

• You must deeply understand your audience before doing keyword research

Most teams jump straight to keyword research. However, Rebecca emphasized the importance of starting with GA4 data, audience affinities, and psychographics to build accurate personas that shape content, UX, and messaging.

• Diversify keyword strategy using long-tail, life-moment, and product-specific queries

Instead of chasing volume, focus on terms tied to user behavior and unique selling points (USPs). Identify when to direct users to PDPs vs PLPs and look for subcategory gaps to expand visibility.

• Collaborate across teams to align SEO with brand and conversion goals

Keyword and audience insights should feed content, digital PR, social media, and UX. This approach led to +600 AI Overview appearances and 160% increased subcategory revenue growth YoY for one client.

3. Clicks don’t pay the bills: Use this audit framework to prove content revenue – Helen Pollitt

Key takeaways:

• Conversions don’t always equal revenue

Not all conversions are equal; some customers refund, others become loyal. Without understanding lifetime value, many SEO reports overstate actual business impact.

• Use a three-step content value framework: ROI, assisted value, and classification

Helen explained how to calculate organic ROI by subtracting hosting costs from revenue per page. Then factor in assisted conversions and cross-channel influence to fully classify a page as high, medium, low, or net negative.

• Cross-team collaboration is key to accurate value assessment

To get the full picture of content value, you’ll need to speak with developers (for infrastructure costs) and finance teams (for conversion value assumptions) to align SEO with company financial metrics.

4. Last-click attribution is dead: Here’s how to fix it – Luke Carthy

Key takeaways:

• Use GA4’s attribution paths report to get the complete picture

The attribution paths report in GA4 reveals how different channels contribute at early, mid, and late stages. It highlights SEO’s often invisible influence throughout the funnel.

• Improve tracking with user IDs and server-side implementation

Features like user_id in GA4 and server-side tracking, connect user behavior across devices, reduce data loss from cookies, and surface more complete data.

• Rebuild SEO attribution with custom GA4 audiences

Luke shared a method for creating GA4 audiences based on users who started their journey via organic search and purchased within a custom time window. This approach reveals a much higher ROI from SEO, in one case, showing £49k in attributed revenue vs. £16k using last-click alone.

5. U-SERPed: AI features & how to appear in them – Tom Capper

Key takeaways:

• No vertical is safe anymore

Some sectors with low AIO presence earlier this year are now heavily impacted. The AIO rollout was rapid and widespread across industries and intent types.

• Brand Authority is a stronger predictor of AIO presence than Domain Authority

Tom’s data shows that branded search volume and Brand Authority metrics now correlate more with AIO visibility than traditional link-based metrics like Domain Authority.

• You can’t outrun the AIOs, but you can choose how to respond

Businesses have two paths: resist the change by focusing on featureless queries, or embrace the change and treat SEO as a brand marketing channel measured by impressions, not just clicks.

• Measurement needs to evolve alongside the SERP

With AIOs, tracking by position or domain isn’t enough. Tom suggests a brand-first approach to measurement, focusing on mentions, presence, and depth within AI responses rather than traditional rankings.

6. How to launch, grow, and scale a community that supports your brand – Areej AbuAli

Key takeaways:

• Trust comes before creation

Most communities fail because they’re treated like marketing campaigns, not relationships. According to Areej, visibility is a byproduct of trust and must be earned before building communities. 

• Use the launch, grow, scale framework

Areej’s three-phase model starts with finding your “why,” defining your culture, and choosing a platform only after those foundations are clear. Growth focuses on retention, onboarding, and ambassador programs. Scaling involves gathering insights, handling change, and regularly revisiting your mission.

• Operational support is critical but often overlooked

Communities can’t be an afterthought dumped on a single marketer. They need resources, cross-team collaboration, and ongoing feedback loops to thrive. Otherwise, engagement and value quickly dry up.

7. How to dominate in AI search in 2025 – Charlie Marchant

Key takeaways:

• AI search visibility hinges on reviews, citations, and digital PR

LLMs like ChatGPT and Gemini pull heavily from third-party sources like review sites and Reddit threads. Citations are the new backlinks to optimize for generative search.

• AI search is personalized but still influenceable

 While AI responses vary by user and context, they’re still driven by rankings, crawlability, and website quality. Clean technical SEO and optimized content remain foundational.

• Local SEO plays a surprising role in AI chat results

LLMs surface map results and local business listings from Google Maps, Bing Maps, and review platforms. Local SEO is more relevant than ever for service-area businesses to increase visibility.

8. The new content playbook you need for success post AI – Chima Mmeje

Key takeaways:

• Create content that forms habits and captures demand

I argued that affinity is the future of content marketing. Affinity leads to more direct traffic, citations, and conversions. To build it, you must be first to market with original content that didn’t exist before you created it.

• No links, no problem. You can compete on SERPs with a new content experience

Some brands have figured out how to rank for highly competitive keywords without backlinks. Using Better Trail as an example, I explained how you can outrank major brands with intuitive UX, personalized filters, and first-person content. 

• Use repurposing to power multimodal content

I explained how to use webinars as a base for multimodal content and lead generation. The ecosystem helps you meet users across channels while reinforcing the same value prop, all built from a single piece of content.

9. Brand and SEO sitting on a tree: K-i-s-s-i-n-g – Lidia Infante

Key takeaways:

• Search engines and LLMs learn about your brand from everything

Every brand mention on Reddit, Crunchbase, or a 2015 blog post is part of the AI training set. Consistency and clarity across your web presence are crucial, even for outdated platforms like old social profiles.

• Build and maintain a consistent entity profile

Use schema markup, unify your brand name and tagline across all platforms, and ensure the accuracy of knowledge panel sources like Wikidata, Wikipedia, and Crunchbase. Even favicon updates and Open Graph tags help reinforce brand identity to machines.

• Repurpose strong content where AI gets its data

Audit your top-performing pieces and repackage them for Reddit, YouTube, LinkedIn, and Quora—platforms commonly cited in AI Overviews. Lidia also shared how SurveyMonkey uses old blog content to appear in third-party search results without traditional rankings.

9. How to build AI tools that automate your SEO workflows– Andy Chadwick

Key takeaways:

• Use AI to write and run code

Tools like ChatGPT, Claude, and Google Colab let you build and run Python scripts in-browser. Andy demoed a content prioritization tool he created in 75 minutes, using SEO API tools like Moz and OpenAI API to score keywords by conversion likelihood.

• Level up by adding clustering, categorization, and APIs

Once your script works, it’s easy to extend it. Andy explained how to plug in Keyword Insights’ clustering API and add categorization logic to group and prioritize thousands of keywords quickly.

• Turn your script into a slick web app with Streamlit

Streamlit transforms ugly code into user-friendly tools. It even hosts apps for free. You can ask ChatGPT to convert your Colab script into a Streamlit app and deploy it without needing a server or tech team.

• Avoid common mistakes: prompt properly, protect secrets, debug smartly

Andy advised the audience to write prompts like spec docs, hide API keys, and not to trust AI to fix bugs. Use error logs, Google, and other LLMs to resolve issues and keep iterating.

What’s happening next for MozCon?