In partnership with
Training cutting edge AI? Unlock the data advantage today.
If you’re building or fine-tuning generative AI models, this guide is your shortcut to smarter AI model training. Learn how Shutterstock’s multimodal datasets—grounded in measurable user behavior—can help you reduce legal risk, boost creative diversity, and improve model reliability.
Inside, you’ll uncover why scraped data and aesthetic proxies often fall short—and how to use clustering methods and semantic evaluation to refine your dataset and your outputs. Designed for AI leaders, product teams, and ML engineers, this guide walks through how to identify refinement-worthy data, align with generative preferences, and validate progress with confidence.
Whether you're optimizing alignment, output quality, or time-to-value, this playbook gives you a data advantage. Download the guide and train your models with data built for performance.
🗣️ Top Stories
If you work in marketing, product, or tech - and want to keep pace with the shifts AI is driving across search, software and strategy - here’s what you need to know from the week just gone.
💼 Oracle–OpenAI’s mega compute pact stuns markets –
TechCrunch reports a surprise $300bn, five-year Oracle compute deal for OpenAI - a move to reduce single-vendor dependency on Microsoft. A fast-follow from Reuters said Microsoft and OpenAI have a non-binding framework for revised terms (details pending), while Reuters (citing The Information) reports Microsoft will use Anthropic’s Claude for some Microsoft 365 Copilot features - hedges on both sides.
Key takeout: This is mutual de-risking. OpenAI is going multi-cloud; Microsoft is going multi-model. With the MS–OpenAI pact only non-binding for now, expect rapid routing changes under the hood over the next few weeks.
⚖️ Penske (Rolling Stone, Billboard) sues Google over AI Overviews –
Reuters says Penske Media has filed a federal lawsuit alleging Google’s AI Overviews use its journalism without permission, killing traffic and affiliate revenue. It’s the first major U.S. publisher challenge focused squarely on AI Overviews, with Penske citing Google’s search dominance as coercive.
Key takeout: Following the recent DMGT legal challenge, pressure on Google and its AI summaries is escalating—a real test of whether “answer engines” can quote journalism at scale without providing compensation.
💸 Google brings ads into AI answers worldwide –
The Decoder reports Google is embedding ads directly inside AI-generated results, with “AI Max” in global beta across Google Ads surfaces - positioning Google as both answer publisher and ad seller.
Key takeout: Although not 'new news' (Google has been experimenting for a while), monetisation is moving away from the “ten blue links” to the generated answer itself - a structural shift with obvious implications for publishers and performance budgets. Earlier in the month Axios reported that in the U.S., AI search ad spend is projected to jump from about $1bn in 2025 to $25.9bn by 2029 (0.7% → 13.6% of search), while total search spend rises from $148.6bn to $190.7bn.
I have finally been able to publish the super insightful discussion that I had with Louise Linehan, of
Ahrefs, that we recorded over the summer.
Watching back, it is crazy to think how much has changed since we spoke. AI Mode has rolled out in many more regions and languages, OpenAI released GPT-5 and I launched a new company called
Obsero.
Louise is brilliant. She is a very well-respected content marketer who has produced some amazingly insightful articles that have helped to broaden our understanding of how AI Search works and what marketers need to know.
‘An Analysis of AI Overview Brand Visibility Factors’ is a piece that she wrote that got a lot of traction and I wanted to get in a little deeper and understand how the Ahrefs content engine works and how Louise thinks the space is evolving.
Make sure to check out the video discussion between me and Louise further down.
I also featured on Trade Informer with Mike regarding Obsero, how information discovery for customers is evolving and what the future of search holds. Check it out!
Thanks! Enjoy the week ahead.
Andy
📖 What I’ve been reading this week in business, marketing & tech
👇 Tap-worthy reads worth your time.
✨ AI
🔍 Google & Search
📲 Growth
Louise Linehan on building brands in the age of AI search
AI search is upending the usual SEO metrics: impressions rise while clicks fall. In the middle of that storm sits Louise Linehan, content marketer at Ahrefs - one of the sector’s most used SaaS platforms.
Having worked her way through Pi Datametrics, BuzzSumo and now Ahrefs, Louise brings a perspective that blends scrappy start-up energy with enterprise discipline. When she writes her content, the SEO world listens.
A history of SaaS enterprise
Louise has gained a lot of traction and attention for her insightful approach to organic growth. That experience shaped how she thinks about product, data and storytelling.
BuzzSumo’s bread and butter was content analytics, tracking which stories gained traction online. “We considered Ahrefs one of our closest competitors,” Louise recalls. “Their Content Explorer mirrored our content analyser. We even had a mentions tracker, like their Alerts. There was always overlap.”
What BuzzSumo lacked in scale, it made up for in pace. A 20-person team, constantly shipping and iterating. That scrappy rhythm - ship, test, refine - was exactly what she saw reflected at Ahrefs. “I liked the way they worked. It was a natural fit.”
Inside the Ahrefs content engine
Fast forward a year and Louise now sits within Ahrefs’ content marketing machine, reporting into Ryan Law, who in turn reports to CMO Tim Soulo. Each marketer is targeted with creating and refreshing content monthly. Ideas are pitched, tested against traffic potential, and then moved through a pipeline with editorial oversight and design support.
The standards are high. Frameworks like “bottom line up front” ensure every article leads with clarity. Another is “mutually exclusive, collectively exhaustive” - no gaps, no overlaps. Outlines go through Ryan before drafts are written. And every piece is tracked, both for performance and quality.
AI plays a role, but not in the way some may think. Louise is open about the balance: “We’ve tested wholesale AI articles. With the right process and editing, they can work. But the vast majority of our content is human-written, sometimes AI-assisted. It’s about using the tools intelligently, not pressing a magic button.”
🎯 Hey content creator!
Thinking about launching your own newsletter? I recommend Beehiiv - it’s built for creators, with all the tools you need to grow:
✏️ Beautiful templates
🚀 Powerful growth features
📈 Built-in analytics
💸 Monetisation tools
All in one place, no fuss.
🐝 Try Beehiiv with my link
(You'll help support me if you sign up via this link 💙)
Cracking AI visibility
Her most high-profile research at Ahrefs has focused on AI Overviews - and how brands get cited. The standout finding? Brand mentions strongly correlate with inclusion in AI answers.
“The data showed what many suspected,” she explains. “The more your brand appears online, the more likely you are to feature in AI results. But I wouldn’t want people to see that as an excuse to stuff their name everywhere. Quality and context matter just as much as volume.”
This is a warning to those that love the hacking mentality. The research encourages marketers to build their brand in a meaningful way rather than taking shortcuts, such as guest posting sprees, shoehorning into listicles, engineered mentions. Louise’s take is blunt: “Those shortcuts will get nuked. You might as well invest in building your brand sustainably.”
👋 Got 30 seconds?
If you’ve got half a minute spare over your morning espresso, I’d really appreciate you taking 30 seconds to fill in this short survey.
No weird personal questions - just a few quick ones to help me better understand who’s reading this, so I can keep improving and work more efficiently with the right partners.
Bonus: I may run a small prize draw with one respondent chosen at random.
Take the survey
Where does Louise sit in the SEO, AEO, GEO debate?
Few debates have polarised the industry like the arrival of terms such as GEO (Generative Engine Optimisation) and AEO (Answer Engine Optimisation). Are they genuine evolutions or just rebranded SEO fundamentals? Louise leans towards the latter.
Her latest research found 76% of AI Overview links already rank in the top 10 search results and 86% rank in the top 100. “From a data perspective, strong SEO fundamentals remain crucial. Chunking content, adding schema, optimising entities - we’ve done these for years. The names might change, but the principles are the same.”
That doesn’t mean nothing is different. Question-led content, FAQ formats, brand pages and controlling crawler access are gaining new importance, which of course includes measurement. But for Louise, the idea of SEO being renamed isn't valid: “Most of it is still SEO.”
Measuring brand in an AI world
How to measure in an AI Search world?
If visibility is no longer just about clicks, how should marketers measure it? Ahrefs’ answer is Brand Radar - a tool tracking how often brands are mentioned or cited across AI assistants. It’s a step towards a new “share of voice” metric for AI search.
Louise offers an example. In one prompt about SEO tool pricing, Ahrefs was described as credits-based - a model abandoned over a year ago. The citation came from an old Reddit thread. “That’s when you realise AI visibility is also brand reputation management. If your brand is misrepresented, that shapes perception in ways traffic metrics won’t show.”
The bigger point: declining clicks don’t mean failing SEO. The role of the organic growth specialist now stretches into customer research, surveys, CRM integrations and sentiment monitoring. It’s a more nuanced, cross-functional brief.
The brands that will succeed
Risk, creativity and the qualitative edge
For all the talk of metrics, Louise is adamant that the brands who thrive in AI search will be those willing to gamble. “Great marketing is risky as hell,” she says, referencing a colleague’s article. “We’ve leaned too heavily on quant. AI opens up new queries, new contexts. The brands that embrace qualitative insights, community building, thought leadership - they’ll be remembered.”
This is perhaps the biggest mindset shift. AI search is not just about being seen; it’s about being trusted. And trust is won through creativity and consistency, not just rankings.
What’s next for Ahrefs
On the roadmap: they have already evolved from a SEO tool to full marketing platform. Expect new social media tools, expanded AI visibility tracking across ChatGPT, Perplexity, Copilot and Gemini, and the ability to monitor custom prompts. “We want to help marketers improve all of their visibility, not just in search,” Louise notes.
It’s an evolution that mirrors the industry’s direction - from keywords and SERPs to brand sentiment, answers and new ecosystems.
Remind me about this Franc Talking newsletter
One click to tell your AI assistant to remember this issue featuring Louise Linehan of Ahrefs on the evolution of AI search, why brand visibility is crucial, and how SEO fundamentals remain vital.
