Last year, roughly 1.2 million people were laid off across industries. Marketing and business development aren’t immune, and in fact, Deb Andrews argues these functions are being innovated faster than most. Add in the reality that we’re now more than three years into generative AI, and the message in Episode 3 of Candid Conversations: Marketing and AI Unscripted is clear: the window to get started is still open, but it’s closing.

In this episode, Deb Andrews (President and Owner of Marketri) and Brady Lewis (Senior Director of AI Innovation at Marketri) cover three topics:

1. How professionals at every level can stay relevant in the AI era
2. What AI means for websites and search
3. A hands-on look at Perplexity Pages, a tool that transforms dense reports into interactive dashboards in minutes.

Staying Relevant: The Gap Is Real, and It’s Growing

Brady opens with an important distinction: what’s happening with AI isn’t like previous technology shifts. Past technologies changed and eliminated jobs, but gradually — in slow, manageable waves. AI is different. It’s happening in large chunks, quickly, and the psychological weight of feeling like you might be replaced compounds everything.

His reassurance is grounded but honest: we’ve survived every major technology shift before, and this one likely won’t be different. The exception is velocity, and velocity is exactly the problem for anyone still sitting on the sidelines.

Brady breaks the current landscape into three groups:

• A large group that hasn’t done anything with AI
• An equally large group that has played with it briefly but never gone deeper: no training courses, no consistent use
• A small group that has been digging in and using AI close to its actual capabilities. This is the group that is starting to move out of reach

That third group isn’t just getting more efficient. They’re learning faster because of AI, using it to build skills in prompt engineering, automations, and new tools at a pace that compounds over time. As Brady puts it: “The longer you wait, that’s more and more ground that you have to make up at some point in time if you want to stay on the same pace as your colleagues.”

Deb adds the workforce reality that makes this urgent for business leaders: a new college grad who knows how to use AI well might perform 90% of the tasks a seasoned professional handles, but at 90x the output. That’s not a distant hypothetical. That’s the hire your competitors are making right now.

Her advice to anyone still hesitating: stop framing this as optional. AI is embedded in the tools you already use. HubSpot has AI features, Google search now surfaces AI overviews, and the frontier models are increasingly how research gets done. If you’re still defaulting to Google blue links for everything, you’re already operating at a disadvantage.

Marketri Academy: Building the Path from Tools to Automation

Wanting your team to be innovative is one thing. Building a structured way to actually get there is another. Deb shares that Marketri is launching Marketri Academy, a 12-month internal program designed to move their marketers from competent tool use all the way to understanding automations and basic agent creation.

The goal isn’t to turn every marketer into a developer. It’s to give them enough context to walk into a client situation, recognize where AI-driven automation or agent workflows could create real value, and consult intelligently on that opportunity.

What’s striking is that the program does not start with AI tools, but with strategic thinking. The rationale is sound. What makes every AI interaction more valuable is something a machine can’t replicate: the pattern recognition, judgment, and lived experience you bring to it. Build that foundation first, then layer the technology on top.

The program itself is homegrown rather than outsourced. We bring in third-party articles, guides built with Claude and ChatGPT, and internal videos from Brady on prompting best practices. It’s a cost-effective model other organizations could realistically adapt.

Brady surfaces a challenge that any technical leader will recognize. People can’t strategize around capabilities they don’t know exist. If your team doesn’t know that A can connect to B, they won’t think to build something where A connects to B, even if the technical barrier to doing it has been removed. That’s why the technical liaison role matters not just as a builder, but as a teacher of what’s possible.

The bottom line from this section is simple. Get busy on your AI journey. If you’re already on it, stay on it. And if you lead a team, ask yourself honestly whether you’re giving people a real pathway to upskill, or simply telling them AI matters and leaving them to figure it out.

Websites in the AI Era: GEO, LLMs, and the End of the Page-Count Metric

Marketri is currently rebranding and undertaking a full website development project. When they went to market, reaching out to roughly 10 development companies found through LLM searches, they hit a wall: almost no one had a concrete strategy for building websites optimized for AI search. Not just vague assurances of “building in an optimized way,” but actual thinking about how LLMs interact with content. That gap forced Marketri to build their own best practices framework.

SEO vs. GEO: Not the Same Thing

Brady introduces the key concept GEO, or Generative Engine Optimization. Where SEO optimized content for Google’s search algorithms (heavy on keyword density, long-form text, topic repetition), GEO is about optimizing for how LLMs actually read and evaluate content. And they work very differently.

LLMs aren’t scanning for keywords. They’re reading your content and quickly deciding whether it answers the question being asked. That changes what good content structure looks like:

• Chunked text instead of long walls of copy
• Descriptive H2 headers that signal what each section covers so an LLM can decide whether to read further
• Structured content like how-to steps and FAQs that can be parsed and cited quickly
• Speed of comprehension as the governing principle: if an LLM can’t read and evaluate your content fast, it moves on

Brady’s analogy makes it concrete. Just as Google historically penalized slow-loading pages, LLMs penalize content that’s hard to parse. The standard has shifted from “do you have the keywords” to “can I understand you instantly.”

The Bigger Shift: Is the Multi-Page Website Already Obsolete?

Deb raises a more provocative question, if the goal of a website used to be getting visitors to click through multiple pages and spend time browsing, does that metric even matter anymore? Her vision of the ideal website isn’t measured in page views. It’s a single, visually impressive experience built around a chatbot that surfaces exactly what each visitor needs.

Brady agrees and takes it further. We’re already seeing AI agents take action within websites on behalf of users. And the LLM companies themselves (ChatGPT with its new browser, Comet with its own) appear to be building toward owning the entire browsing ecosystem. The strategic question isn’t just how to optimize your website for AI. It’s whether individual websites will be competing with AI browsers that handle everything within their own apps, with tools like Photoshop now built directly into ChatGPT.

As Brady puts it, buckle up. The changes coming in 2026 and beyond will arrive faster than anything we’ve seen before.

Tool Spotlight: Perplexity Pages

Perplexity is a retrieval augmented generation (RAG) tool, meaning it pulls information from LLMs rather than being a large language model itself. It’s already a go-to for a certain kind of research. But the feature Brady highlights is something different: Perplexity Labs.
Within Perplexity, users can select the Labs option alongside standard conversation and deep research modes. Labs gives Perplexity the ability to write and build code, including fully functional web pages, directly within the app.

Here’s how Marketri used it in practice:

• Deb had generated a list of roughly 60 website best practices using Claude
• Brady took that document, dropped it into Perplexity Labs, and prompted it to create an interactive dashboard
• In about four to five minutes, while you can watch it write the code in real time, Perplexity produced a fully functional HTML dashboard
• The result: 51 items organized by category, converted into an interactive checklist with progress tracking, export functionality, and a shareable link

That dashboard has since been used across 10 client calls. The difference between handing a client a six-page static document and walking them through an interactive, categorized checklist is significant, both in how it lands and in how useful it is in a meeting.

What makes this particularly relevant is the accessibility. Brady describes it as “one of the easiest ways right now for a non-technical person (such as a marketer) to do something that really stands out.” No code required. No developer needed. Just a prompt and a file.

Key Takeaways for Marketing and Business Leaders

• If you haven’t started your AI journey, you are behind. That’s not meant to invoke fear. It’s meant to create urgency. Three years into generative AI, the time for a wait-and-see posture has passed.
• The people learning AI are learning faster because of AI. The gap between adopters and non-adopters isn’t growing linearly. It’s compounding.
• Strategic thinking comes before AI tools. Marketri Academy starts with strategy, not software. The judgment and pattern recognition you bring to AI determines the quality of what you get out of it.
• GEO is not the same as SEO. LLMs read and evaluate content differently than search algorithms. Chunked text, descriptive headers, and structured content like FAQs and how-to steps are the new standard.
• Perplexity Labs is worth exploring. If you have a dense report that needs to be usable in a client meeting, this is one of the fastest ways to turn it into something interactive, with no technical skills required.

Watch previous episodes of Candid Conversations or explore our AI and marketing resources for more insights on navigating this transformation.