AI, GEO, and How to Take Advantage of the Shift (Edition #7)

Riley Kearl - Senior Marketing & Sales Growth Partner | B2B Funnel Clarity & Pipeline Systems – Turn Traffic ​into Revenue


April 30, 2026



Welcome back to Edition 7 of Funnel Clarity Weekly! The one after this will include the big free release (Edition 8 funnel clarity tool + definitions sheet). This week we're talking about something I've been watching closely for a while now: AI, GEO, and why the companies moving slowly on this are making the same mistake we saw with mobile back in 2015.


What AI actually means for your marketing and sales funnel

When people say "AI" they mean a lot of different things. Worth noting we've been using AI for years without calling it that: think algorithms, recommendation engines, spam filters. In B2B specifically: CRM lead scoring, programmatic ads, and chatbots. The current wave is just more visible, more accessible, and moving faster. And you don't need to be a tech company or an AI company to feel the impact. You do need to think about two things though: how you show up externally in AI, and how you use it internally.

1. External: how AI affects the way customers find and experience you

This is where it connects directly to your funnel and where most B2B companies are moving too slowly.

AI agents visiting your site: this one is already happening whether you like it or not. AI agents are visiting your website right now on behalf of human users and pulling your content to serve tools like Claude, ChatGPT, Perplexity, and Gemini. You may not see these visits clearly in GA4 yet but they are happening. Your website is becoming a source document as much as a destination. The question isn't whether AI agents are visiting (they are). The question is can they easily find what they need when they get there?

GEO (Generative Engine Optimization): this is your response to the above. GEO is about optimising your content so that when AI tools generate answers, they cite and reference you rather than a competitor (or at least alongside them). AI agents are visiting your site regardless; GEO is what you do to make sure what they find is worth surfacing to the humans asking the questions. We'll dig into some practical steps below.

AI product integration: if you're a tech or SaaS company, AI is likely being built into products across your category right now. A lot of companies are doing this to be able to say they "have AI." Some of it is genuine, some of it is fad. Your customers will figure out the difference quickly. If you're not a tech company this one may not apply to you directly, but understanding where your category is heading is still worth your attention. Either way, AI is affecting your company.

2. Internal: how mature is your AI usage as a team?

You don't need to be at Agentic to get value, start where you are and build from there.

Efficiency: AI assists you and your team individually day to day. Using ChatGPT to draft emails, Claude to summarise long documents, Perplexity to research a topic quickly. Think of AI as your always-on assistant that never sleeps and never complains. This is the right place to start if you haven't dabbled in AI much yet.

Workflows: this is where individual efficiency becomes team efficiency. You start building repeatable processes that incorporate automation so things happen consistently without someone manually triggering each step (but you will likely have human reviews along the way).

Agentic: using multiple LLMs together in automated workflows that make decisions using your actual company context: your ICP, personas, values, past sales data, and financials. This is where AI gets genuinely powerful and is still early for most companies. Human review is still important.

The progression from Efficiency to Workflows to Agentic is a journey, not a switch you flip. Most B2B teams are at Efficiency right now. Getting to Workflows is a realistic near term goal. But it's worth starting to think about what Agentic could look like for your business specifically. Building out a new GTM motion based on past results and custom inputs? Creating proposals or targeting plans based on a prospect's unique requirements, past successes, and the deals most likely to win? That's where this is heading, we’ll explore some examples below.


So what is GEO and why does it matter for your funnel?

GEO stands for Generative Engine Optimization. I described it in Edition 3 as "advanced SEO" and that's still the best way I can explain it. The goal is the same: show up when your buyer is looking for what you offer. The difference is where they're looking. Some call it AEO, AIO, LLMO … different acronyms, same idea.

More and more B2B buyers are starting their research in AI tools rather than search engines. They type a question into Claude or another AI tool and get a summarised answer (sometimes with sources). If you're not one of those sources, you don't exist at that moment.

Remember from Edition 2: traffic comes to your site from humans, AI agents, and search crawlers. AI agents are visiting your site on behalf of human users and serving your content back to them through tools like Claude, ChatGPT, Perplexity, and Gemini. You might not see the human visit in GA4 but your content is still being surfaced (or not) depending on how well structured it is.

Your website still matters. AI still needs sources to reference. But the path from awareness to your website is changing, it’s like the 2015 mobile traffic shift all over again.


What you can realistically do right now

This doesn't need to be a massive project. Here's where to start:

1. Get the basics in place

Make sure your site has a robots.txt file (tells crawlers what to index) and a sitemap.xml (shows search engines/AI Agents your URL structure). This matters for SEO and GEO. While you're in there check that your robots.txt is not blocking AI crawlers. Some older setups block anything they don't recognise and AI agents like GPTBot (OpenAI), ClaudeBot (Anthropic), and PerplexityBot fall into that category. If they're blocked you're invisible to them.

2. Create an llms.txt file

Similar to robots.txt but specifically for AI agents. The llms.txt gives them the key information about your business: what you do, who you serve, your pricing, your offering. This is so they can more easily find it and serve it accurately to human users. It’s like a briefing document for AI. Here's a good example of what one looks like: data.guavy.com/llms.txt

3. Create a dedicated AI agents page

A public page (not linked in your main navigation) built specifically for AI bots to read and understand your offering, use cases, and pricing. Clear headings, no jargon, answer-first structure. Think about how a robot would like to see a page structured. Here's an example: data.guavy.com/ai-agents

4. Structure your content for AI

AI tools prefer content that is easy to parse and cite. That means clear headings (H1, H2, H3), short paragraphs, bullet lists, comparison tables, and FAQ sections. If a human can scan it in 10 seconds, an AI can too. Start with 2-3 pages: a FAQ page, a comparison page, and a "best for" page that answers specific questions your buyers are actually asking. This has already been talked about under the AEO (Answer Engine Optimization) banner for years, just accelerated by AI. Surface the answer! Give me the answer! (say both humans and AI Agents)

5. Third party “citations” matter

The majority of brand mentions in AI answers come from third-party sources, not a brand's own website. Brands are significantly more likely to be cited through third-party content than their own domain, with most of those mentions coming from listicles, comparison pages, and review roundups. It means getting listed on review sites and industry directories (Concurate). Or on platforms that are most cited by AI tools like YouTube, Reddit, Wikipedia, Forbes, Quora, and Linkedin (Niel Patel).

6. Content freshness

AI engines weigh recency when selecting sources, a guide published in 2024 with no updates will lose ground to a more recent article on the same topic. Adding "Last updated" timestamps and refreshing cornerstone content regularly helps.

7. Category definition content

Articles that define your category are among the highest-priority GEO investments for B2B SaaS brands. When AI systems explain what a category is, they cite the most comprehensive, authoritative definition. Writing the definitive "what is X" content for your category is a strong GEO play. (Gupta Deepak)

8. Track AI referral traffic in GA4

Most people don't know this is possible yet. You can set up tracking to see traffic coming from AI referrals. Simple but powerful.


Worth revisiting: your website and AI traffic

Even if fewer humans visit your website directly, AI agents are still reading it on their behalf. Your website is becoming a source document as much as a destination. So while you might see website traffic plateau or shift in GA4, your content is still being read/used, just not always by human eyeballs.

This is why I keep saying in this series: your website still matters. It's where AI is referencing from. The humans might be getting the answer in ChatGPT but ChatGPT got it from your site.


The more advanced version: multi-LLM workflows

Skip this section if you're still at the basics stage. But for those thinking ahead, here's where AI gets genuinely exciting for B2B marketing and sales.

Instead of using one AI tool for one task, you start combining them in automated workflows where each LLM does what it's best at. Here's a real GTM workflow example:

You provide the inputs: target country, ICP, platforms to use, company documents, past sales data. Then the workflow runs: ChatGPT identifies target companies → Perplexity creates and refines the messaging → Gemini verifies and double checks → emails are sent.

The whole process runs with minimal human intervention (though human review at key stages is still important) and gives you something actually useful and specific to your business rather than generic results. This is what tools like N8N are built for: connecting multiple AI tools into a single automated process.

A few other examples of what Agentic could look like for your business:

Most B2B teams aren't here yet, but start asking yourself: what would an ideal agentic process look like for your business? Write them down and come back to them as you progress in your AI journey. You may need someone to help you here, I’m still learning myself and would not consider my processes agentic yet either, but it’s worth a conversation to see where you’re at and where you want to be.


Try this today

Run these starter prompts in ChatGPT, Claude, Perplexity, and others … as results vary by platform. What comes up? Are you there? Use the same prompts and track them over time:

Note whether you appear accurately (2 points), vaguely or inaccurately (1 point), or not at all (0 points). Add up your score across all three platforms. That's your GEO baseline, run it again in 3 months. Message me for an expanded prompt sheet if you want to take this to the next level past the three above.

If the multi-LLM workflow section got you curious, N8N is a good place to start experimenting, it's free and well documented.


Next up is Edition 8 and the full release. The Funnel Definition Sheet, the interactive Funnel Clarity Tool, and the full workshop process outline. All free for subscribers.

If you've been following along since Edition 1, thank you. It was fun resurfacing the post I wrote in 2015 about mobile: same pattern of thinking, different technology. As Stephen King (not that one) commented back then: “technology shifts always offer opportunity to companies who seize the day!” Still true.

Edition 8 is going to be worth the wait. See ya soon.

Riley


BONUS: I wrote about this type of change before. In 2015.

About 11 years ago(!) I wrote an article called "The Amazing Mobile Advantage for Your Small Business." It was about the rise of mobile web traffic and Google's mobile-specific algorithm changes. How people were accessing the internet shifted and Google was rewarding sites that were optimised for it. Reading it back I reminisced about my early marketing days, but also realized that some of it could have been written just yesterday.

Here's what I wrote in July 2015:

"So I know you are happy the way things are, why should you care? Why go through all the effort when Google will change things again anyway? Technology is moving quickly, you need to be proactive. You need to keep up with algorithm changes in general. So what about mobile? Is it just another crazy Google initiative?"

Now let’s swap the words to make it relevant to today:

"So I know you are happy the way things are, why should you care? Why go through all the effort when the way people search will change again anyway? Technology is moving quickly, you need to be proactive. You need to keep up with AI changes in general. So what about LLMs and GEO? Is it just another crazy tech initiative?"

And this one from the same 2015 article:

"If you fall behind in mobile web optimization it will be hard to catch up. Tech moves fast and it will only get faster. Therefore, you should have a plan in place and take the steps towards improving website usability on smartphones."

How I would update it for today:

"If you fall behind in AI adoption and GEO it will be hard to catch up. Tech moves fast and it will only get faster. Therefore, you should have a plan in place and take the steps towards improving the way you show up in AI."

Almost word for word. Back then the shift was in where people browsed: from desktop to mobile web. Today the shift is in where people search: from traditional search engines to AI tools. Different tech medium, same mistake, same types of companies making it. Slow adoption, resistant to change, comfortable with the status quo.

I'm not saying to panic. I'm saying start paying attention and start to take some initial steps towards embracing mobile…ahem I mean AI and GEO.

(Read my 2015 mobile post here which includes links to the original article/link via Wayback Machine)


Read Edition #8 on LinkedIn