AI Won’t Hit Your Number: Why the Tools Can’t Replace the Judgment

You run marketing at a company that’s scaling faster than your headcount. The number went up again this quarter. Your team is already at capacity. And the loudest advice in your inbox says the answer is obvious: point AI at the gap. Agents that plan campaigns, write the briefs, build the funnels, run the tests. A lean team punching like a big one.

Here’s the part nobody selling you the tools will say out loud: the tools are real, and they still won’t hit your number. Not because they’re weak. Because hitting a number was never an execution problem.

It’s a judgment problem, and judgment is the one input agentic AI can’t supply.

If you’re a VP or Head of Marketing right now, that’s the actual tension you’re living in. You don’t need someone to tell you AI is coming — you’ve already deployed half of it. You need a clear read on where it helps, where it quietly hurts, and what stays yours no matter how good the models get. That’s the read this post is for.

The uncomfortable research finding

In 2023, Harvard Business School researchers ran a field experiment with Boston Consulting Group — 758 consultants, real work, GPT-4 in the loop. On tasks that sat inside what the models are good at, the results were dramatic: consultants using AI finished more than 40% higher quality work than the control group, faster and in greater volume. That’s the headline everyone quotes, and it’s true.

But the researchers did something smarter. They also gave people a task designed to sit just outside the model’s competence — a business problem that looked AI-friendly but required reconciling data the model would misread.

The authors called this the “jagged frontier.” AI is spectacular on one side of a line and quietly terrible on the other, and the line is invisible from where you’re standing. (Dell’Acqua et al., “Navigating the Jagged Technological Frontier,” 2023; now in Organization Science, 2025.)

Read that finding as a marketing leader and the lesson is sharp: the value AI created wasn’t in the doing. It was in knowing which side of the frontier you were on. That knowing is judgment. And the people who had it got a force multiplier; the people who didn’t got faster, more polished mistakes.

Why “AI won’t hit your number” is a strategy sentence, not a tech one

Your number is a chain of decisions before it’s ever a chain of tasks. Which segment is actually worth winning this year. Which problem your product solves that a competitor can’t copy back. What “a good lead” means, precisely enough that sales and marketing stop arguing about it. Where the next dollar compounds versus where it just gets spent.

An agent can execute any campaign you point it at. It cannot tell you the campaign was aimed at the wrong buyer. It will optimize a funnel toward a conversion that doesn’t turn into revenue, and it will do it beautifully, at scale, without ever flagging that the goal was wrong. The model has no opinion about what’s worth doing. It only has velocity. Velocity toward a bad aim is how teams end a quarter exhausted and behind.

This is why the “will AI replace marketers” debate is aimed at the wrong altitude. The routine production layer — first-draft copy, variant generation, audience pulls, reporting — is genuinely getting automated, and you should let it. The layer that sets the number is not: deciding what to build a market around, reading what a buyer actually wants versus what they say, choosing what not to do when everything looks urgent. That layer is scarcer now, not less scarce. When execution gets cheap, the quality of the decision in front of it is the whole game.

A framework: the keep / automate / supervise cut

You don’t need a philosophy of AI. You need a decision rule you can apply to your own stack by Friday. Sort every marketing activity into three buckets by asking one question: what happens if this is wrong and nobody notices?

Run your current AI deployments through that cut this week. Most leaders find they’ve automated the right things, under-supervised the middle, and quietly let the tools creep into the keep-it-human layer because it felt productive. Pulling judgment back up to where it belongs is usually the single highest-leverage move available — and it costs nothing but a decision.

What this means for a lean, scaling team

The counterintuitive part: adopting AI well makes senior marketing judgment more valuable to you, not less. The tools flatten the cost of production, which means the differences between teams stop being about output and start being almost entirely about aim. Two companies can run the same stack, the same agents, the same volume — and one compounds while the other spins, purely on the quality of the decisions steering the machine.

That’s the real reason a scaling company hits a wall it can’t tool its way past. Not a shortage of execution capacity — you’ve got more of that than you’ve ever had. A shortage of the one input the execution capacity can’t generate for itself. The team can do anything now. The open question is whether anyone senior enough is deciding what’s worth doing.

Where a fractional CMO fits

If you’re a marketing leader carrying more than you can personally direct, the gap AI opened isn’t a hiring gap — it’s a judgment-bandwidth gap. You don’t necessarily need another full-time executive above you. Sometimes you need a senior sounding board: someone who has sat in the chair, who can look at where you’ve pointed the machine and tell you which decisions are on the wrong side of the frontier before you’ve spent a quarter finding out the hard way.

That’s what a fractional CMO does in an AI-era marketing org — not more hands, sharper aim. Which is exactly what the tools can’t sell you.

Sources: Fabrizio Dell’Acqua, Edward McFowland III, Ethan Mollick, et al., “Navigating the Jagged Technological Frontier: Field Experimental Evidence of the Effects of Artificial Intelligence on Knowledge Worker Productivity and Quality” (Harvard Business School Working Paper, 2023; published in Organization Science, 2025).

superwired is a fractional CMO practice helping founder-led and growth-stage companies build marketing that’s wired for the AI era — real strategy and senior leadership, without the full-time cost.