Article's Content
Article's Content
For the last 15 years, I was satisfied with my ability to explain SEO to my friends and, more often than not, the clients who hired me (or the agency I worked for) to execute SEO on their behalf.
I knew where I stood and how to give the elevator pitch.
SEO worked. Search-optimized content plus optimized marketing funnels equaled business growth. The sun shone, the grass grew, and everything was gravy.
But with Generative Engine Optimization, things are no longer gravy. Decidedly not gravy.
GEO is changing the way people make decisions. Zero-click search is kicking us in the proverbial, and my winning strategy, which still drives ranking positions, no longer drives comparable growth.
However, LLMs like ChatGPT are also an opportunity. The businesses that most effectively execute strategies that drive LLM visibility and citations will be the ones that succeed in this new normal.
So I’m dusting off the ol’ keyboard and writing an article about how I’m thinking about it, and what I’m now recommending to my clients.
Let me walk you through it.
GEO = SEO (ish)
Before we dive into all the ways you need to change everything you’re doing and throw the baby out with the bathwater, I need to make one thing clear: Traditional SEO still matters. The high-ranking content on Google is what gets picked up and cited by LLMs.
(For the record, high-ranking content still drives significant traffic, but let’s put that to the side for now.)
Generative answers (what ChatGPT says when you ask it a question) are pulled predominantly from the top-ranking URLs, so, at least in terms of citations, investing in SEO keeps paying off.
Two things ruin this for everyone:
- LLMs are conversational (unlike Google). They blend citations, paraphrase sources, and pull from multiple sources to create their answer. So, even if you’re in the top position, if your competitor is in second, the LLMs will also pull from their dumb opinion alongside yours.
- Zero click. Being the content or domain cited by the LLM means you get to inform the answer, but it doesn’t necessarily mean you get traffic. In fact, if your answer is great, that only makes the LLM’s answer great, and users don’t need to do anymore research (into, for instance, your actual website). This is equally true in the AI Overview, of course.
Nonetheless, the point remains: optimizing your site and content for search = optimizing your site and content for LLM visibility, share of voice, and citation metrics.
That brings us to the real distinction:
SEO is your space: your website, blog, technical optimization, etc.
GEO is all that stuff plus external influences: the other sources of the LLM’s answers. And, some of those, you can actually do something about.
Beyond SEO: 5 Ways to Influence LLMs
I’ll say it again: GEO is about doing all the SEO stuff you’ve been doing, plus expanding into the environments models learn from and pull their answers from.
Here are five sources (beyond your domain) that LLMs learn from and you can influence:
1. Reddit
LLMs love Reddit. A) It ranks (yes, still!) and B) It’s messy, opinionated, and rich with what the LLMs perceive as authentic human language. After all, it’s what they were trained on in the first place…
Actionable Strategy: Get involved in real discussions. Engage in relevant threads. Foster community engagement.
That “engage in relevant threads” is a great one, by the way. GEO tools like Profound can help you identify the top-cited Reddit threads (the posts ChatGPT is pulling answers from). Adding yourself to those conversations is a no-brainer.
DO NOT be spammy. The goal isn’t to force links; it’s to build credible context around your brand and expertise.
2. Placements in High-Ranking URLs
If LLMs are citing Forbes Advisor, TechCrunch, or niche industry publishers like ProjectManager, then being featured there means your brand gets pulled into AI-generated answers.
Actionable Strategy: Partner (pay for a relationship) with the publications that already dominate Page One. Sponsorships, advertorials, and bylined pieces all feed the generative ecosystem.
3. LinkedIn Native Publications
LinkedIn has quietly become a treasure trove for LLM training data. Expert-driven, high-engagement posts are particularly sticky.
Actionable Strategy: Publish consistently. Think in frameworks, insights, and commentary that models can learn from. Each thoughtful post helps teach the machine who the expert is (you, in case there was confusion).
4. Review Sites (Capterra, G2, TrustRadius, etc.)
When someone asks an LLM for “the best CRM” or “the top AI writing tools,” the model pulls from aggregated review data alongside Reddit posts and “pay-to-play” listicles.
Actionable Strategy: Treat your review profiles like SEO assets. Solicit reviews, ensure they’re up to date, and activate happy customers. In the GEO era, reputation literally trains the algorithm.
5. YouTube
Video transcripts are searchable data, and they’re increasingly part of model training pipelines. Tutorials, reviews, and explainers all feed the web’s collective understanding.
Actionable Strategy: Create helpful, keyword-rich videos that mirror how users actually phrase their prompts. The right phrasing makes your content retrievable to both people and the LLMs.
What GEO Actually Optimizes For
We don’t really optimize for generative engines; we influence them. A successful GEO strategy shapes the data layer that trains and informs large language models.
LLMs are pattern-recognition systems. They infer credibility from recurrence, context, and association. That means your brand presence needs to be persistent and contextually coherent across all the sources that matter: Reddit threads, editorial mentions, Q&A threads, product reviews, YouTube videos, etc.
How to Measure GEO Success
Traffic and keyword positions tell only part of the story now. GEO success looks more like share of synthesis. How often your brand is referenced, paraphrased, or surfaced in generative answers.
Top metrics include:
- LLM Visibility: Are your URLs or brand mentioned in AI outputs (ChatGPT, Perplexity, Gemini, etc.)? Are you being mentioned in the right answers?
- LLM Citations: When an LLM answers a searcher’s question, are they pulling from your domain (or Reddit threads you authored) to get that answer?
- LLM Sentiment: When you’re mentioned in the LLM’s answers, are you discussed positively, or are threads and reviews with negative sentiment impacting what they’re saying about you?
- Referral traffic from LLMs: This one is straightforward. If you’re cited and people click on the link, track that traffic (GA4 does this reasonably well).
And, of course, standard SEO metrics still matter. They’re part of how you measure GEO success (see “GEO = SEO(ish) above).
For more on the ROI aspect, I discussed it at length with Foundation’s brilliant in-house writer, Ethan Crump, which resulted in this piece: How Brands Should Think About the ROI of Generative Engine Optimization.
The Future of GEO
We’re still in the early innings of generative search. Soon, new optimization tools will track how models interpret, cite, and compress web data.
But, at the heart of it, GEO doesn’t replace SEO. It builds on it.
The same instincts (and strategies) that helped SEOs like me succeed in 2010 — great content, proven best practices, the communication of credibility and an optimized domain — still matter.
There are just a few more things to throw on the strategic pile:
- Create once, distribute widely (fine, Ross… Distribute “Forever”): Each new platform that LLMs ingest from (pay-to-play listicles, Reddit, LinkedIn, YouTube) is another opportunity to teach the model who you are.
- Speak the model’s language. Alongside schema markup, LLMs understand clarity, structure, and consistency. Write like your ideas need to be learned, not just read.
- Build trust cues. Reviews, citations, backlinks, and mentions all reinforce credibility. The more human validation, the more machine confidence.
- Stay current. Models retrain and refresh. Outdated content fades. Freshness is authority.
So keep publishing. Keep optimizing for the things you have been. But most importantly, stay active in shaping the narrative that the machines will tell about your brand.
If you’re rethinking how your brand shows up in the generative search era, Foundation’s GEO Strategy team can help. James and his team work with B2B brands to build visibility frameworks that drive search visibility and AI citations.
