Key Concepts
- AEO (AI Engine Optimization): The practice of optimizing content and online presence for visibility within AI-powered search and answer engines like ChatGPT, Gemini, and Perplexity.
- LLMs (Large Language Models): AI models that can understand and generate human-like text, forming the basis of tools like ChatGPT.
- JavaScript Rendering: The process by which web browsers execute code to display dynamic content. Many LLMs do not render JavaScript, impacting how they access information.
- Atomic Units of Content: Breaking down content into small, focused paragraphs or sections, each addressing a single topic, to improve LLM comprehension.
- Repetition and Consensus: The strategy of ensuring a key message or data point appears consistently across multiple online platforms to build credibility with AI models.
- New Loop Marketing Framework: HubSpot's marketing framework that includes AEO as part of its "Amplify" phase.
AEO: Optimizing for AI Surfaces
The discussion centers on the emerging field of AI Engine Optimization (AEO), defined by Mike King as "the art and science of getting visibility in AI surfaces like ChatGPT, Gemini, Perplexity, etc." This is crucial as predictions suggest AI-generated traffic could rival or surpass traditional search engine traffic by 2028. The core challenge for marketers is understanding how to optimize for these new channels and prioritize efforts.
Getting Started with AEO
For businesses new to AEO, the first step is measurement. This involves understanding where a brand currently stands in terms of visibility on AI platforms and identifying opportunities where they are not appearing. Mike King suggests prioritizing branded terms, particularly those related to pricing, as a starting point. Historically, some businesses avoided displaying pricing directly on their websites, leading to a loss of control over brand messaging. When AI tools source pricing information from third-party sites, brands miss the opportunity to manage this crucial aspect of their message.
HubSpot has developed a free tool, AEO Grader, launched in March 2024 and available in six languages, to assist with this measurement. It allows users to input their brand and compare their visibility against competitors across major answer engines.
On-Page Content Optimization for LLMs
A significant technical difference between traditional search engines like Google and AI platforms like ChatGPT is their handling of JavaScript rendering. While Google and Bing can interpret JavaScript, ChatGPT and similar LLMs cannot. This means content rendered solely by JavaScript may not be accessible to these AI models.
To address this, HubSpot implemented a strategy of publishing pricing blog posts to make their pricing information accessible. This resulted in an immediate and significant improvement in pricing accuracy across AI services.
Beyond JavaScript, content structure is paramount. Similar to traditional SEO, clear headings, bullet points, and tables are beneficial. However, for LLMs, content should be broken down into "atomic units". This means avoiding paragraphs that cover multiple topics and instead creating focused paragraphs dedicated to a single subject. This makes it easier for LLMs to parse information and use it in their responses.
Furthermore, using clear, ideally unique data points is crucial. If a fact is corroborated by other sources, AI models are more likely to use it. Mike King likens LLMs to "lazy readers," emphasizing that walls of text or multiple ideas within a single section reduce the chances of information being extracted.
What Doesn't Work as Well for LLMs
While some SEO tactics are transferable, others are not. For instance, removing meta descriptions to encourage Google to surface the most relevant copy for a higher click-through rate, which works for classic SEO, is counterproductive for LLMs. In the ChatGPT environment, a meta description that "spoils what the answer is on the page" is more effective. This highlights the need to "put the answer first" and avoid "burying the lead" when catering to LLMs.
Off-Site AEO Strategies
Off-site factors are critical for a successful AEO strategy. Reddit is highlighted as a platform performing exceptionally well. When it comes to link building, the focus has shifted from volume to relevance. Acquiring a few high-quality, highly relevant links is more impactful than obtaining a large number of less relevant ones.
Leveraging Reddit involves building a strategy around identifying relevant threads and subreddits for engagement. Brands can also establish their own subreddits, but this requires thoughtful community management. Inauthentic promotion is strongly discouraged on Reddit; genuine engagement and strategic participation are key.
HubSpot has observed that increased engagement on their own subreddit leads to a rise in positive mentions of HubSpot across Reddit generally, demonstrating a ripple effect.
Digital PR and securing brand mentions are also vital. The strategy involves getting clear messages and data points highlighted in pitches and content. This can involve creating supporting content assets. The goal is to disseminate these messages across as many platforms as possible.
The concept of creating microsites on specific subjects is also being revisited as a way to control messaging across the web. The strategy of repeating a key statistic or message across multiple touchpoints is likened to creating "raffle tickets" to increase the chances of winning visibility in AI models' "raffles." This ensures that when AI models conduct their queries, they repeatedly encounter the desired information.
Synthesizing AEO Strategy
A comprehensive AEO strategy involves:
- Measurement: Understanding current visibility and identifying gaps using tools like AEO Grader.
- Strategy Development: Creating a plan that encompasses both on-site (owned content) and off-site efforts.
- Repetition: Consistently reinforcing key messages and data points across all relevant platforms where AI models are known to source information.
- Alignment: Ensuring all channels work cohesively behind a unified content strategy and foundational message, rather than operating in silos.
The conversation emphasizes that AEO is fundamentally a content strategy problem rather than purely a technical one.
The Last Word: Spicy Takes on Marketing
This segment features quick, opinionated statements on marketing trends.
- Kyle Denhoff: "Marketing teams need to optimize for the overlap." This means identifying the two to three channels where the audience spends time and where LLMs are sourcing information, and focusing foundational efforts there. The principle is to achieve a "two for one" return on investment.
- Aja Frost: "We need to stop saying 'this channel is dead' about any channel." She argues that even seemingly declining channels can be effective if there's a good concept and an audience present. Declaring a channel dead prematurely can lead to giving up before exploring its potential.
- Mike King: "It is a profound mistake to say that optimizing for generative AI is just SEO." He asserts that viewing AEO solely through an SEO lens demonstrates a lack of business understanding and can lead to missed opportunities. A successful AEO strategy requires buy-in from across the business and involves more than just SEO tactics, which are likely to evolve.
Conclusion
The discussion underscores that the rise of generative AI and AI-powered search necessitates a new approach to online visibility. AEO is a multifaceted discipline that requires a deep understanding of how AI models process information, a strategic approach to content creation and distribution, and a willingness to adapt traditional marketing tactics. The key takeaway is that AEO is not simply an extension of SEO but a distinct and evolving area that demands a holistic, content-centric strategy with cross-functional buy-in.
AI summaries can miss context or contain errors. Check important details against the original video.