• by Cory Treffiletti , Featured Contributor, 6 hours ago

I spent the first third of my career in agencies, building the digital media business.  In those days, the org chart was the product.  You sold the intelligence of good people and their ideas. You had account teams, planning and buying teams, creative teams, and a whole layer of management whose job was to keep all those people coordinated. The client paid for the top people.

The agency’s job was to keep the seats full and productive. That was the model, and for a long time, it worked well enough that nobody asked too many questions about it. 

Then came the big consulting shops.

About a decade ago, Accenture, Deloitte, IBM and their peers decided they wanted a piece of the marketing and media business. Their pitch was simple -- and, frankly, very compelling. The consultants could replicate process models across clients at scale, bring media and creative capabilities under the same roof as business consulting, and offer a more accountable relationship than the traditional agency had ever really delivered.

There was pressure and waves of consternation, but the holding companies survived.  They did so by mostly acquiring capabilities, outspending on talent, and leveraging the relationships they had already spent decades building. It was expensive and uncomfortable, but they held the line.

Everyone assumed the next threat would come from the same direction. Something bigger, better-resourced, more institutional.

Whoops.

The real disruption is coming from somewhere they didn’t foresee, and it’s moving faster than anyone in a holding company wants to acknowledge out loud.

Agentic workflows are collapsing the infrastructure advantage that protected the big shops for 30 years. A five- to 10-person agency running the right AI stack can now plan, execute, optimize, and report on a media campaign with the same sophistication as a 500-person operation.

I get inundated with messages from small shops proclaiming their capabilities, and while at first I was skeptical, I sat down to think it through and talk with some other old agency peeps. We realized that AI is really a massively unsettling problem for the big shops. 

Ironically, the consultants who thought they had finally figured out how to compete with the agencies are now facing the same threat from below that the agencies faced from them.

The agentic agency model doesn’t just change who can compete. It changes the entire look of the competition. Agentic tools absorb the coordination that is required in an FTE-based agency model.  They are more efficient and dependable, and they can handle more processes simultaneously, with human oversight ready to review and engage with customers. 

This opens the door to a commercial model that is genuinely better for both sides. If you are no longer selling full-time equivalents, you need a different structure, and the most logical one is performance-based pricing, where the agency participates in the upside of what it produces.

This is the model that aligns incentives: The agency wins when the client wins. The work is measured by outcomes, not by effort. For smaller shops that can now execute at the level of the big players, this is not just a pricing adjustment; it’s a weapon.

The downside is that, in an agentic system that optimizes in real time, the pressure to deliver results compresses quickly. Clients will want to see the numbers in weeks, not quarters. The problem is, real brand building does not show up in a 30-day ROAS window. If performance-based pricing defaults to short-term metrics because those are the ones the agent can produce fastest, we will have automated our way back into a challenging market for the agency model.

The agencies that survive this moment will not be the ones with the most people or the largest holding company behind them. They will be the ones with the clearest point of view about what they are actually selling. Agentic tools level the execution playing field for everyone: the boutique shop, the mid-size independent, and yes, the consulting firm that thought it had finally learned to speak the language of media. The pressure is on brands to determine how much leash they’ll give the agencies they employ in this new paradigm of advertising services.