Adding AI or headcount won’t fix a marketing skills gap

Your team asked for two more people this budget cycle. Finance came back with an AI license and expected the same team to do more. Six months later, the team produces more drafts than ever, and your best person spends much of each week fixing them.

Adding people or AI to a team whose skills haven’t caught up with the work produces more of the wrong work, faster. Before the AI money went out, nobody asked a basic question: Is this team short on skills, short on people, or short on both? Those are different problems with different fixes, and AI responds differently to each.

A skills gap is the distance between what the work requires and what your team can do in production with its tools. Usually, the team learned the job the stack used to require. Vendor training stopped at email scheduling before it reached behavioral scoring, or AI tools showed up, and nobody taught anyone how to write a brief that produces usable output, or how to check what comes back.

A headcount gap means the team knows how to do the work and runs out of hours. Say your team runs 12 campaigns at a time, at full quality, and the plan calls for 40. You can streamline approvals and automate reporting, and throughput goes up a little each time, then stalls well short of 40. Getting there takes program managers, content producers, and analysts you don’t have. A training course can’t close that gap.

Mixing the two up sends money to the wrong fix. Train people who are already maxed out, and they have no hours to use what they learned. Hire into a team that lacks the skills, and the new people learn the same workarounds everyone else uses.

What AI does to each gap

AI muddies the picture because it gets sold as both. One pitch says it does the work of an analyst. The next says it’s extra hands. Budgets suggest CMOs aren’t treating it as a replacement for people. In Gartner’s 2026 CMO Spend Survey, the research and advisory firm’s annual survey of 401 marketing leaders, labor’s share of the marketing budget rose from 21.9% in 2025 to 24.5% in 2026. The top barrier to AI-driven efficiency, cited by 38%, was a lack of internal AI expertise and talent.

“AI is not a shortcut around marketing capability,” said Ewan McIntyre, who leads Gartner’s marketing research.

Put AI on top of a skills gap, and you get more output at the same quality, plus a new job: checking it. A survey of 2,003 marketing leaders, commissioned by Optimizely, which makes website content management, A/B testing, and personalization software, and conducted by research firm Savanta, found that 76% spend at least three hours a week editing, fact-checking, or correcting AI output. More than half (54%) said leadership underestimates the human effort it takes to make that output usable. MarTech senior editor Constantine von Hoffman covered the study in July.

Checking the output takes skill, too. Someone who doesn’t understand segmentation can’t catch a bad segment, no matter how fast a tool built it. So AI raises the skill bar for whoever signs off on the work.

It also eats the hours. In the same study, 39% said they’re too busy managing workflows and outputs to think strategically, and 46% worry heavy AI use will erode skill development among junior marketers. That second number matters for anyone counting on AI to cover a skills gap. The tool can slow down the learning that would close it.

Put AI in the hands of a skilled team that’s short on hours, and you get a different result. People who know what good work looks like can review output quickly, so the time AI saves stays saved. It can take a share of the volume off people who already do the work well. The team running 12 campaigns still needs program managers to reach 40.

How to tell which gap you have

Three checks get you most of the way.

Look at what gets fixed

Ask the people reviewing the work, AI-generated or not, what they correct. Fixes about judgment, such as the wrong audience, customer records matched to the wrong accounts, or an off-brand claim, point to skills. Fixes about rushed details, such as typos, missed deadlines, and skipped steps, point to hours.

Look at where work waits

If work sits in a queue for the one person who knows how to set up a journey or build an audience, you have a skills gap concentrated in one person. That’s also a headcount risk because when that person leaves, the skill leaves with them. If work waits because everyone is busy and anyone on the team could do it, you’re short on hours.

Give the team time and watch what happens

Pull one recurring project off the team’s plate for a month. If the quality and volume of their other work go up, they were short on hours. If they turn out the same work at the same quality, they were short on skills, and more time won’t change that. The test costs you a month of that project, so pick one you can afford to pause.

Recommended Articles

Why AI is complicating marketing centralization

Nov 5, 2026

How to turn AI productivity gains into marketing ROI

Nov 5, 2026

Your CEO wants answers your dashboard can’t give

Oct 23, 2026

When it’s both, fix skills first

Plenty of teams are short on skills and people at the same time. When that’s your team, fix the skills first. The two gaps feed each other. A team short on skills burns hours on rework and workarounds, so a skills gap often looks like a headcount gap. Fix the skills, and part of the headcount gap goes away.

Once the team knows how to do the work, every hire and every AI license you add after that makes the team stronger. This order has a real cost. Training takes months, and the volume targets don’t pause while people learn. You’ll probably miss some of this year’s output goals to hit next year’s. Most teams can’t fund both fixes at once anyway. In the same Gartner survey, 56% of CMOs said they lack the budget to deliver their 2026 strategy, and 54% reported insufficient resources.

You can work on both gaps at once in one way: Make your next hire a skills hire. Bring in one experienced person who knows the work and whose job includes teaching it. You get an extra pair of hands now and a better-trained team six months from now.

For the budget, the benchmarks I laid out in “The martech investment nobody budgets for” still hold: training at 15% to 20% of annual license cost, and one process and operations role for every three to four core platforms. Apply the same rule to AI tools. A $100,000 AI license comes with a $15,000 to $20,000 training line.

Before the next AI request goes to finance, add one line to it: which gap this purchase closes, skills or people, and how you know. When the answer is unclear, run the three checks first. The AI tool will still be for sale next quarter.