I Use AI Every Day as a Fractional CMO. Here's What It Still Can't Do.

I built an AI system this morning before my Earl Grey was cold.

Not a prompt. A system. I have custom skill files that feed my brand voice into every piece of content I produce. I use AI to research competitors, build outlines, draft blog posts, structure email sequences, and synthesize campaign data across channels. I've built AI-powered content workflows that would have taken a junior marketer two weeks to produce manually. I do it in an afternoon.

I'm telling you this because I want you to understand that I am not the person who's going to tell you AI is overhyped. I use it more than a lot of CMOs I know, and it has fundamentally changed how I work. It's made me faster, more thorough, and more productive than I've ever been.

And every week, at least one founder asks me some version of the same question: "If AI can do all that, do I even need someone like you?"

Fair question. Here's what I've learned.

The $99 illusion

A startup launched a product this month they're calling an "AI CMO." For ninety-nine dollars a month, it promises to deploy a team of AI agents to handle your SEO, content, and growth marketing. Automatically. No marketing leader required.

They're not the only ones. The market is flooded with tools promising to replace marketing leadership with automation. AI strategy platforms. AI content engines. AI campaign managers. The pitch is always the same: why pay for a marketing leader when a subscription can handle it?

I get the appeal. If you're a founder at $2M ARR, bootstrapped, stretched thin, spending 10 hours a week approving blog posts and managing an agency that isn't delivering, the idea that you could hand all of that to a $99/month tool sounds like the answer to a prayer.

Here's what actually happens.

You subscribe. The tool starts producing content. A lot of content. Blog posts appear. Social posts get scheduled. Maybe some ad copy gets generated. Your content calendar, which was empty three months ago, is suddenly full. It feels like progress.

Except your pipeline doesn't move.

The blog posts are answering questions nobody in your market is asking because no one with strategic context told the AI what questions matter. The social posts sound like every other SaaS company's social posts, because AI writes to the average by default. The ad copy is grammatically perfect and strategically empty because the AI doesn't know your positioning, your competitive landscape, or what your sales team hears on every call that doesn't close.

You're not doing marketing now. You're doing more of what wasn't working before, just faster.

I've walked into companies that look exactly like this. Three months into their AI marketing experiment, they have a Google Drive full of content, a social calendar running on autopilot, and a team that's more confused than ever about why none of it is translating to revenue.

The content wasn't the problem. The missing strategy was.

AI is a force multiplier. That's the problem.

Here's the reframe most people miss: AI doesn't replace marketing strategy. It amplifies whatever you already have.

If you have a clear ICP, sharp positioning, and a channel strategy built on real data, AI makes you dangerous. It accelerates content production, speeds up research, and frees up your time for the strategic work that actually moves the needle.

If you have none of that, AI just helps you produce noise at scale.

A force multiplier multiplies force in whatever direction you're already pointed. If you're aimed at the right target, you hit it faster. If you're aimed at nothing, you just miss faster.

The companies replacing marketing leadership with AI tools aren't saving money. They're spending it faster on the wrong things.

What AI actually handles well (and what I use it for daily)

I want to be specific here, because vague claims about AI capabilities are part of the problem. Here's what I actually use AI for in my fractional CMO practice, and why it works:

Content production. AI cuts my first-draft time by roughly 60-70%. I use it to generate blog post drafts, email sequences, social content, and case study frameworks. But here's the part that matters: I'm feeding it brand voice guides, research briefs, and the strategic direction I’ve built. The AI isn't deciding what to write. I am. It's executing against a strategy that already exists. Without that strategic input, AI content is just...noise. It fills space. It doesn't fill the pipeline.

Research and competitive analysis. AI can synthesize a search landscape, pull competitor messaging, analyze content gaps, and summarize what's ranking for a target keyword cluster faster than any human researcher. But it can't decide what to do about what it finds. The gap between "here's what the market looks like" and "here's what we should do about it" is a strategy gap. AI lives on one side. Experience lives on the other.

Data synthesis. Pulling patterns from campaign performance, summarizing metrics across channels, and flagging anomalies in conversion data is genuinely impressive with AI. But the analysis layer isn't where decisions get made.

AI can tell you that your email open rates dropped 15% last month. What it can't tell you is that the drop coincides with a messaging shift your sales team also noticed on calls, that the shift happened because you repositioned after a competitor launch, and that the real problem isn't email performance at all. It's a positioning gap that's showing up everywhere. That connective tissue between data points and business context is where the actual value lives. AI sees patterns. It doesn't know why they exist. A marketing leader who's been in the room, heard the sales calls, and understands the competitive landscape can tell you why in a conversation. AI will give you a chart.

These are real capabilities. They matter. And they've let me move faster as a fractional CMO. AI hasn't threatened my role. It's upgraded it.

But notice the pattern: in every single case, AI is executing against strategic decisions a human already made. The brand voice guide. The research brief. The channel strategy. The positioning framework. Remove the human strategic layer, and you're left with a very fast tool that doesn't know where to aim.

Where AI falters

Strategy and prioritization. AI can generate a marketing plan. Ask it to build a 90-day growth strategy, and it'll produce something with the right section headings, reasonable-sounding tactics, and enough buzzwords to look credible in a board deck. What it can’t do is decide which plan is right for your business, at your stage, and with your constraints. Strategy isn't about knowing what could work. It's about knowing what will work here and now, and more importantly, what to say no to. AI has no opinion about what to cut. A good CMO knows where not to spend.

Positioning and differentiation. This is where AI fails most quietly, and most expensively. AI is trained on everyone's messaging. It's absorbed thousands of websites, pitch decks, and value propositions. When you ask it to write your positioning, it gives you a weighted average of what already exists. The output sounds professional. It also sounds like your three closest competitors. Positioning is about being specific and different. AI defaults to safe and similar. 

Pattern recognition from experience. When I walk into a company with a stalled pipeline, a frustrated sales team, and a junior marketer who's working hard but flying blind, I know what's wrong within two weeks. Not because I'm smarter than AI. Because I've walked into that exact situation dozens of times across different industries, company sizes, and market conditions. That pattern recognition isn't in a training dataset. It's in 20 years of sitting in the seat, making the calls, and seeing what happens. AI can pattern-match across text. It can't pattern-match across lived experience.

Organizational context and politics. Knowing that the VP of Sales has never trusted marketing. Knowing that the CEO is terrified of committing budget because the last agency burned $100K and delivered nothing. Knowing that the junior marketer is talented but has never had a strategic leader to learn from, and what she needs is direction, not replacement. AI can't read a room. It can't navigate the politics of a leadership team. It can't earn the trust that makes a founder willing to hand over a function they've been white-knuckling for three years.

Accountability. When the strategy isn't producing results in month two, someone has to make the call: do we adjust the messaging, shift the channel mix, or give it another month? That's a judgment call made by someone who owns the outcome. AI doesn't own outcomes. It doesn't lose sleep when pipeline is down. A marketing leader does.

The real danger: AI-accelerated random acts of marketing

Here's what connects all of this: without strategic direction, AI just automates the problem I see most often in growth-stage companies. I've written before about the hidden costs of random acts of marketing. The pattern is always the same: activity without architecture. Campaigns that don't connect to a strategy. Tactics that don't tie to revenue. Spend that looks productive and produces nothing measurable.

AI makes random acts of marketing faster. That's it.

A recent Gartner survey found that 65% of CMOs expect AI to dramatically change their role in the next two years, but only 32% say significant changes are needed to the CMO skill set. That disconnect is revealing. The leaders who think AI changes the game but not the skills are the ones who haven't yet realized that AI amplifies the need for strategy. It doesn't reduce it.

The companies that will win in the next two years aren't the ones that adopted the most AI tools. They're the ones that paired AI with real strategic leadership. AI will handle the execution layer while a human with experience, judgment, and accountability directs the whole system.

The bottom line

AI made it cheaper to produce marketing. It didn't make it cheaper to produce the right marketing. That still takes someone who's done this before.

If you're committed to running marketing with AI tools, do one thing first: build the strategic foundation the AI needs to work from. Your ICP. Your positioning. Your messaging framework. Your channel priorities. A brand voice guide that actually sounds like your company, not like a SaaS Mad Lib. 

That foundation is something I build for companies in a few weeks, documented and structured, so your AI tools have something to actually execute against.

If your marketing feels like a lot of AI-generated activity with not much to show for it, that's not a technology problem. Let's talk about what's missing.