AI Will Take Some Marketing Jobs. Here’s Which Ones.

AI will replace some marketing jobs. Not all of them, not most of them, but some, and probably more than the optimists are currently willing to admit. The more honest question is not whether displacement will happen, but which roles are genuinely at risk, which are not, and what separates the two categories.

After two decades running agencies and managing large marketing teams, my read is this: AI is not coming for marketing as a discipline. It is coming for the parts of marketing that were never really about thinking in the first place.

I have been watching AI reshape the agency model in real time, and the pattern is not subtle. The roles disappearing are not the ones that require the most experience. They are the ones that require the least judgment.

What AI Can Already Do in Marketing

Before the job displacement conversation makes any sense, you need an honest picture of what AI can actually do well right now, not in theory, not in a vendor demo, but in practice.

AI can write first drafts of copy at scale. It can generate social media variations, summarise research, produce SEO-optimised blog structures, write product descriptions, draft email sequences, and pull together performance reports from raw data. It can do all of this faster and cheaper than a junior team member. Semrush has a useful overview of where AI is already embedded in marketing workflows, and the list is longer than most people expect.

AI can also handle a growing slice of paid media management. Automated bidding, audience segmentation, creative testing, and budget allocation are all areas where machine learning has been doing the heavy lifting for years. Google’s Performance Max campaigns are the most visible example, but the logic applies across platforms.

What AI cannot do reliably, at least not yet, is exercise commercial judgment. It cannot read a room. It cannot tell you that the brief is wrong. It cannot look at a campaign result and understand why it happened in the context of what was going on in the business at the time. It produces outputs. It does not own outcomes.

That distinction matters more than any other in this conversation.

If you want a broader picture of how AI is being applied across the marketing stack right now, the AI Marketing hub at The Marketing Juice covers the landscape in depth, from tools and tactics to strategy and measurement.

Which Marketing Roles Are Genuinely at Risk

Let me be specific, because vague reassurances about “human creativity” do not help anyone make a career decision.

Junior content writers producing volume output are at significant risk. If your primary job function is writing a certain number of blog posts per week to a brief, AI already does that faster and at lower cost. The question is not whether the tool can match your quality on a good day. It is whether the business can afford to keep paying for something it can now get for a fraction of the price.

Basic SEO execution roles are under pressure. Keyword research, meta description writing, on-page optimisation checklists, and content briefs based on search volume data are all tasks that AI handles competently. Moz has documented how AI tools are already reshaping SEO workflows, and the direction of travel is clear.

Entry-level paid media roles are being hollowed out. The platforms themselves are automating more of what junior media buyers used to do manually. Setting up campaigns, writing ad copy variations, adjusting bids, and producing weekly performance reports are all areas where AI is taking over the repetitive work.

Data and analytics reporting roles, where the job is pulling numbers and formatting them into slides, are at risk. Not analytics strategy. Not insight generation. But the mechanical act of producing reports from data that already exists.

Social media coordinators whose primary output is scheduling and caption writing are also in a difficult position. Buffer has tracked how AI tools are changing social media workflows, and the conclusion is not encouraging for roles built around volume output.

I want to be clear about something. These are not bad people or weak marketers. Many of them are talented. But the roles themselves were designed around doing things that AI now does faster. That is a structural problem, not a personal one.

Which Marketing Roles Are Not at Risk

The roles that are safe are the ones where the value is in the thinking, not the doing.

Senior strategists who can connect marketing activity to business outcomes are not going anywhere. In twenty years of agency work, I have never seen a client pay a premium for someone who could write copy quickly. They pay a premium for someone who can tell them why their current approach is not working and what to do instead. AI does not do that.

Brand and creative directors who set direction and make judgment calls about what is on-brand are safe. AI can generate a hundred creative executions. Someone still has to decide which one is right, and more importantly, which ones would damage the brand. That judgment is not automatable in any meaningful sense.

Client and account leadership roles are safe. The ability to manage a relationship, read what a client actually needs versus what they are asking for, and handle the commercial dynamics of a business partnership is deeply human work. I spent years in those conversations. No AI is replacing that.

Performance marketers who own commercial outcomes rather than just campaign execution are safe. There is a difference between someone who manages a paid search account and someone who is accountable for the revenue it generates. The latter is not at risk. The former is.

Early in my career, I ran a paid search campaign for a music festival at lastminute.com. It was a straightforward campaign by today’s standards, but within roughly a day it had generated six figures in revenue. What made it work was not the mechanics of setting it up. It was understanding the product, the audience, and the timing well enough to put the right offer in front of the right people at the right moment. AI can help with execution. It cannot replicate that commercial instinct.

The Agency Model Is Changing Faster Than Most People Realise

The agency world is where the pressure is most visible right now, and I say that as someone who has spent most of his career in it.

The traditional agency model was built on billable hours. You charged clients for the time it took to produce things. When I was growing a team from 20 to 100 people, a significant portion of that headcount was doing work that could now be done by AI in a fraction of the time. That is not a criticism of those people. It is a description of how agency economics worked.

That model is breaking. Clients are starting to ask why they are paying for 40 hours of content production when AI can produce a first draft in 40 minutes. The honest answer is that they should not be, and the agencies that pretend otherwise are going to lose those conversations.

The agencies that will survive are the ones that shift their value proposition from production to thinking. That means fewer junior staff doing volume work, and more senior people doing strategy, quality control, and client leadership. It also means the junior pipeline gets thinner, which creates its own long-term problem for the industry.

In-house teams are moving more slowly, partly because they are insulated from the commercial pressure that agencies face. But the same logic applies. If a marketing team’s output is primarily content, reports, and campaign execution, AI is going to change what that team looks like.

What Marketers Should Actually Do About This

Telling people to “embrace AI” is not a strategy. It is a platitude. So let me be more specific about what actually matters.

First, understand what AI can and cannot do in your specific role. Not in general. Not in theory. In your actual job, with your actual outputs. Semrush has a practical breakdown of how AI tools are changing content strategy workflows that is worth reading as a reference point.

Second, get good at working with AI tools rather than competing with them. Prompt engineering is a real skill. Knowing how to get useful output from a large language model, and more importantly, how to evaluate whether that output is actually good, is becoming a core competency. HubSpot has a useful guide to choosing between LLMs that gives you a sense of how to think about tool selection.

Third, move toward the parts of your role that require judgment. If you are a content writer, the question is not how to write faster. It is how to develop the strategic and editorial judgment that makes your output worth commissioning in the first place. AI can produce a blog post. It cannot tell you whether that blog post should exist, or what business problem it is solving.

Fourth, get comfortable with data. Not just reading dashboards, but understanding what the numbers mean in a business context. Moz’s research on AI-generated content is a useful read for understanding where AI output tends to fall short, and by extension, where human judgment still adds clear value.

When I started out in marketing, I asked my MD for budget to build a new website. The answer was no. So I taught myself to code and built it myself. That was not a story about coding. It was a story about not waiting for permission to develop a capability that the business needed. The same logic applies now. The marketers who will be fine are the ones who treat AI as a tool to master, not a threat to wait out.

Fifth, think about your commercial value, not just your technical skills. Can you connect what you do to revenue? Can you make a case for marketing investment in terms a CFO would understand? That kind of commercial grounding is what separates marketers who get cut in a downturn from marketers who get protected.

The Bigger Picture: What AI Actually Changes About Marketing

The most important shift is not about jobs. It is about what counts as good marketing work.

AI has raised the floor. Average content, average copy, and average campaign execution are now cheaper and faster to produce than ever. That means the bar for what gets noticed, what gets shared, and what actually moves people has gone up. You cannot win on volume anymore. You have to win on quality, relevance, and insight.

I judged the Effie Awards for a period, which gave me a clear view of what effective marketing actually looks like when it is done well. The campaigns that won were not the ones with the biggest budgets or the most content. They were the ones where someone had thought carefully about the problem and come up with a genuinely sharp answer to it. AI does not change that. If anything, it makes that kind of thinking more valuable, because it becomes rarer relative to the volume of average work being produced.

There is also a question of trust. AI-generated content is proliferating fast, and audiences are starting to feel it even when they cannot name it. The content that builds real brand relationships is content that feels like it came from a human being with a genuine point of view. Crazy Egg has a useful breakdown of how AI is being used across marketing asset production, which gives you a sense of where the volume is going.

The marketers who will thrive are the ones who use AI to do more of the mechanical work, and spend the time they save on the thinking that AI cannot do. That sounds simple. Most things that are true do.

There is more on how AI is reshaping marketing strategy and measurement across the full AI Marketing section of The Marketing Juice, if you want to go deeper on any of these areas.

About the Author

Keith Lacy is a marketing strategist and former agency CEO with 20+ years of experience across agency leadership, performance marketing, and commercial strategy. He writes The Marketing Juice to cut through the noise and share what works.

Frequently Asked Questions