The Zombie CMO

We're Back. And We're Here to Kill the Zombies.

Last week, a CMO asked me about personalization.

This sounds reasonable until you learn their website converts 0.4% of traffic to marketing qualified leads. Not opportunities. Not pipeline. Marketing qualified leads. And yes, they’re still using MQLs in 2026, which is its own red flag.

Their MQL-to-opportunity conversion? Less than ten percent.

But sure, let’s talk about personalizing the experience for the 99.6% of visitors who don’t convert because that’s going to do the trick.

Here’s what I wanted to say but didn’t: “You’re already dead. You just don’t know it yet.”

Welcome to the era of the Zombie CMO.

The Maturity Model You’re Not Seeing

Jordan and I recently presented in Montreal to a room full of rev-ops folks; generally more sophisticated than your average marketing team. We showed them what’s possible with Claude Code and our pain-based segmentation methodology.

We absolutely blew their minds.

We also got zero business from it.

Why? Because there’s a massive gap between “holy shit that’s amazing” and “I know what to do with this.” Most organizations are still stuck in Stage 1 of the AI maturity model: one-off tasks in ChatGPT.

Here’s the actual progression:

Stage 1: One-Off LLMs (where 95% of you live) You use ChatGPT to polish a blog post. Claude helps you brainstorm subject lines. Each task starts from zero. No memory. No compounding benefit. Just parlor tricks that make you feel innovative.

Stage 2: Workflows (where the cutting edge thinks they’ve arrived) Clay tables. Zapier chains. N8N orchestrations. Repeatable, deterministic processes. This is better! Except it’s brittle. When your data source changes or the market shifts, someone has to manually rebuild everything. You’ve automated the assembly line, but you’re still running an assembly line.

Stage 3: Claude Code + LLMs (where the future already lives) This is where it gets scary. Jordan uses Claude Code to stitch NHTSA recall data to dealer databases. He generates 1,800 pain-based value propositions in hours—work that would take your team months, if they could even figure out how to do it at all.

This isn’t automation. It’s augmentation with memory, context, and compounding learning.

The brutal truth? You can skip Stage 2 entirely and go straight to Stage 3. But first, you’d have to admit you’re still in Stage 1.

Your Assembly Line is a Museum Piece

Let me describe your organization. Tell me if I’m wrong:

Product Marketing owns positioning and messaging. They spend months on “market segmentation” using Gartner reports and demographic data from ZoomInfo.

They hand this off to Marketing Communications, who polish the language and create the “awareness plan” that nobody outside MarComm understands or cares about.

Meanwhile, Content Marketing runs its own SEO factory, churning out blog posts optimized for keywords that have nothing to do with what Product Marketing says you sell. Their only job is generating cheap traffic, so that’s what they do.

Digital Marketing runs paid social and PPC campaigns to generate expensive traffic, using messages that sort of align with positioning but definitely don’t align with what Content is saying.

Then everyone hands their leads to SDRs, who discover that nobody actually wants what you thought you were selling because buying is organic and emotional and your entire funnel presupposes rational, linear behavior.

Sound familiar?

This is 1920s production-line thinking applied to a fundamentally non-linear, deeply emotional buying process. You’ve built an assembly line to manufacture funnel conversions that don’t actually exist.

Because here’s what you won’t admit: your buyers are emotional decision-makers who use their analytical capabilities to rationalize monkey-brain choices they’ve already made. Your beautiful funnel metrics are fiction. Your MQLs are fiction. Your attribution models are fiction.

But hey, at least you can report on them in your QBR.

The Organization That Replaces You

The companies that will eat your lunch aren’t doing this.

They’re smaller. Way smaller. And exponentially more capable.

They’re full of what we call GTM Engineers; people who are simultaneously product marketers AND content creators AND campaign builders. They don’t think in channels. They think in methodologies.

One person now owns: market research → pain-based segmentation → messaging → campaign deployment. End to end. No handoffs. No relay races. No MQL definitions.

They use Claude Code to compound context and learning across everything they do. When they discover something in market research, that insight immediately flows through to messaging and campaigns. When a campaign performs, they understand why because they built the whole thing.

Each of these people does the work of five traditional marketers.

And this is before autonomous agents arrive. That’s the next extinction event.

These organizations aren’t coming. They’re already here. You’re competing against them right now, while you’re stuck in a meeting about whether to count a whitepaper download as an MQL.

The Test

Quick diagnostic. Be honest:

• Is your org chart organized by channel instead of outcome?

• Does work move between teams like a relay race?

• Does your “AI strategy” mean someone uses ChatGPT for content?

• Do you still believe funnel metrics predict pipeline?

• Are you asking about personalization when your conversion rates are broken?

If you answered yes to any of these: You’re the zombie.

I know this because I’m the a zombie enabler too. I’m a fractional CMO working with multiple clients stuck in Stage 1, helping them navigate Stage 2, while Jordan is already living in Stage 3. I see the future every day, and then I go back to organizations that are still arguing about MQL definitions. FML.

The cognitive dissonance is staggering.

Wake Up or Get Replaced

This isn’t a warning about the future. The future arrived while you were in your MQL review meeting.

Your board will figure this out—probably right after they replace you with someone who doesn’t ask about personalization when the conversion rate is 0.4%.

Or you wake up. You burn down the silos. You stop organizing by channel. You start building GTM Engineers who think in methodologies and compound their learning with every project.

You embrace Claude Code not as a tool but as a fundamental shift in how marketing organizations operate.

We’ll be writing a lot more about how to make this transition. About what the maturity model actually means. About how to build teams that compound instead of handoff.

For now, just know: the window is closing.

The zombies are already dead.

The question is whether you’re going to stay that way.

What’s Next

Here’s what we’re doing about this: Cannonball is evolving to meet people where they are on the maturity curve. For our GTM engineering community (you know who you are) we’re going deeper on Claude Code, agent frameworks, and the technical methodologies that let you compound your work. For growth leaders trying to navigate this transition, we’re creating a roadmap: how to recognize where your team actually is (not where you think they are), how to build confidence in making the leap, and how to structure organizations that can actually execute at Stage 3. We’re not abandoning the technical community that got us here—they’re the ones showing us what’s possible. But we’re also done watching brilliant, hardworking executives get replaced because nobody explained what’s actually happening. Consider this the opening shot. More coming weekly. The window’s closing, but it hasn’t closed yet.