Do Marketing Agencies Need to Brace for Impact from AI — or Adapt Calmly?

There’s a lot of noise around AI right now. Depending on who you listen to, it’s either going to wipe out marketing agencies entirely or magically fix every problem clients have ever had. As usual, the truth likely sits somewhere in the middle.

From where I’m standing, AI isn’t something agencies should panic about — but it is something they need to take seriously. This shift is real, it’s moving quickly, and pretending it’s a passing trend is probably the riskiest move of all.

The good news? AI doesn’t threaten all agencies. It threatens very specific ways of working.

Who will AI really threaten?

AI is brilliant at speed, pattern recognition, and repetition. It’s getting better by the week at tasks that are rules-based, predictable, and execution-heavy. Things like basic content generation, resizing assets, first-draft copy, simple reporting, even entry-level strategy templates.

If an agency’s value is built almost entirely on doing rather than thinking, churning out volume with little differentiation, then yes, there’s a genuine threat. Not because AI is “better”, but because it’s faster, cheaper, and doesn’t get tired.

That doesn’t mean those agencies disappear overnight. But it does mean margins get squeezed, expectations change, and clients start asking harder questions about value.

Execution-only vs strategic agencies

This is where the gap really opens up.

Execution-only agencies, the ones selling hours, outputs, or deliverables in isolation, are most exposed. If a client can get 70–80% of the same output from a tool, the pressure quickly shifts to price. That’s a race to the bottom no one really wins.

Strategic agencies, on the other hand, are in a much stronger position. AI can support strategy, but it doesn’t own accountability. It doesn’t sit in messy client conversations, weigh up commercial risk, or join the dots across brand, opinions, timelines, and people.

Clients don’t struggle because they can’t make things. They struggle because they don’t know what to make, why, or in what order. That’s still very human territory.

The danger of racing to the bottom

One thing I’m already seeing is agencies rushing to shout “AI-powered” without really thinking it through. Slapping AI on the tin doesn’t automatically create value — especially if it just accelerates low-quality work.

The danger here is agencies undercutting themselves. If you lead with speed and cost alone, you’re training clients to see you as interchangeable. And once that happens, loyalty disappears quickly.

AI should improve quality, clarity, and decision-making — not just output volume. If it’s only being used to do more for less, that’s a short-term win with long-term consequences.

Where agencies should double down

The smartest agencies I work with are doubling down on the bits AI can’t replace easily:

• Discovery and insight – asking the right questions, not just delivering answers

• Creative judgement – knowing what not to do is as valuable as knowing what to do

• Client leadership – guiding, challenging, and protecting outcomes

• Systems thinking – understanding how brand, marketing, tech, and operations connect

• Accountability – owning delivery, not just contributing to it

AI becomes a tool inside these processes, not the product itself. Used well, it frees up time for higher-value thinking rather than replacing it.

Why clients still want partners, not tools

Most clients don’t wake up wanting “AI”. They want clarity, confidence, and results. Tools don’t give reassurance. Partners do.

When budgets are tight or stakes are high, clients still want someone experienced to say: this is the right move, this is the risk, and this is how we’ll handle it if it goes wrong. That level of trust doesn’t come from software alone.

In fact, as tools become more accessible, the role of experienced people often becomes more important — not less. Someone still needs to steer the ship.

New agencies will emerge — and they’ll feel very different

One thing that often gets missed in the “AI will kill agencies” debate is that it’s not just about who survives — it’s about who appears next.

Every big shift creates space for a new type of agency. We’ve seen it before with digital, social, performance, UX, and product-led teams. AI is doing the same thing again, but quicker.

New agencies will emerge that are built with AI from day one. Not bolted on. Not cautiously tested in the background. Properly embedded into how they think, price, and deliver. These teams will be smaller, faster, and far more comfortable blending human judgement with machine speed.

They won’t be obsessed with outputs. They’ll focus on leverage, using AI to remove busywork, compress timelines, and get to decisions quicker. That means more time spent on strategy, creative direction, and client leadership, even with fewer people involved.

They can control the space, if theres hesitation to adapt at all.

Keep a finger on the pulse

All that said, this is changing fast. Anyone claiming to have a fixed answer is kidding themselves. What’s true today will look very different in 12 months from now.

Agencies don’t need to panic, but they do need to stay curious, keep testing, and be honest about where their real value lies. Adapt calmly, invest wisely, and don’t cling to outdated models just because they’ve worked before.

If anything, this shift is a reminder that experience, judgement, and leadership matter more than ever — whether that comes from a senior agency team or trusted specialists brought in when it counts.