Springtime for agencies and service providers in marketing and martech

I had the honor of keynoting the 4A’s Decisions event in Boston this week and spent the day talking with agency leaders about AI. You might expect those conversations to be laced with anxiety — after all, the narrative around AI in the marketing industry has mostly been about disruption, displacement, and existential dread. And sure, there was some of that.

But the dominant energy in the room was something else entirely: excitement.

I think they’re right to be excited. Here’s my hypothesis: while many app-layer SaaS companies are still figuring out where differentiated value lives in an AI-native stack, the opportunity for agencies and service providers has never been greater. More clients need more help than ever before. The tools to deliver that help have never been more powerful. And the structural advantages that services firms have in a fast-moving, uncertain environment are genuinely formidable.

Let me make the case.

In the report I recently published with Databricks, The New Martech Stack for the AI Age, one of the central arguments is that commercial martech is shifting from a world of siloed applications to a world of composable infrastructure. The universal data layer — governed, flexible, and interoperable — is becoming the foundation that everything else builds on.

This shift has enormous implications for agencies and service providers.

In the old model, much of the value in martech was captured by the app layer. Buy the platform, configure it, maybe hire an agency to help with implementation. The app was the star; services were supporting cast. But when the stack becomes infrastructure — a composable canvas of data, compute, and AI capabilities — the app layer opens up. And the opportunity for creative, tailored solutions built on top of that infrastructure gets much, much wider.

Instead of selling clients a siloed application, agencies can plug into a client’s governed data infrastructure and build exactly what that client needs. Bespoke campaigns, custom AI workflows, unique customer experiences. Not cookie-cutter configurations of someone else’s product, but genuinely differentiated work built on a shared substrate.

The surface area for services is expanding dramatically.

Nobody can keep up with how fast AI is moving. Not brands, not vendors, not analysts. (Definitely not this analyst.) The pace of change in AI capabilities, tools, and best practices is genuinely unprecedented.

But service providers have a structural advantage: they work across many clients simultaneously. And that creates an arbitrage advantage — a compounding return on learning and experience that no single brand team can replicate.

When you’re running AI-powered programs and campaigns across a dozen clients in different verticals, you develop pattern recognition that’s incredibly valuable. You see what’s working, what’s overhyped, what’s quietly delivering outsized results. You cross-pollinate ideas, practices, and insights across engagements in ways that would be impossible inside a single organization’s walls. An AI-driven personalization approach that you tested for a retail client informs the next engagement with a financial services firm. An agentic workflow that flopped in one context saves another client from repeating the same mistake.

In a landscape defined by uncertainty and rapid change, this kind of cross-client intelligence is enormously valuable to clients. It’s also a durable competitive advantage for the agencies that cultivate it deliberately. The firms that build systematic ways to capture, share, and apply their cross-client learnings will compound that advantage over time.

At the conference, one micro service provider pulled out his laptop and showed me the team of AI agents he’d built to execute bespoke go-to-market campaigns for his clients. Content generation, audience segmentation, campaign orchestration, performance analysis — all handled by a coordinated squad of AI agents, with humans providing strategy, judgment, and creative direction.

This would have required a service battalion to execute just two years ago.

AI lets small, nimble agencies punch dramatically above their weight class. And in an environment where so much is changing so quickly, nimbleness is a real competitive advantage. Smaller agencies can adopt new AI tools faster, experiment more freely, and pivot more quickly than large firms weighed down by legacy processes and organizational inertia.

Domain expertise intersected with AI expertise is a spectacular combination. A four-person agency with deep knowledge of, say, B2B SaaS marketing and genuine fluency with AI tools can now deliver capabilities that rival much larger competitors.

Now, you might hear “lower barriers to entry” and think race to the bottom. But that’s scarcity thinking. What’s actually happening is abundance. Solutions that would have been economically impractical to build for a single client — too labor intensive, too expensive, too complex to justify — are suddenly within reach. The drop in barriers isn’t just compressing existing work into cheaper buckets. It’s unlocking entirely new categories of work that didn’t exist before.

The age-old analogy that still has legs: when desktop publishing arrived, it didn’t just make existing print jobs cheaper. It created an explosion of new kinds of publications and communications that had never been economically feasible. AI is doing the same thing for marketing services, but at a much grander scale. The total addressable market for creative, data-driven, AI-powered marketing services is expanding, not shrinking.

There’s also a strategic reason clients should be leaning more heavily on services right now: in an environment changing this fast, the ability to experiment quickly has become a competitive advantage.

To be clear, companies still need major platforms and infrastructure underpinning all of this. The composable canvas I described in the Databricks report depends on serious foundational investments in core systems. Those are big, important commitments, and they create the foundation on which everything else depends.

But once that foundation is in place, a growing share of the opportunity shifts to a very different kind of work. Not multi-year implementation programs, but rapid cycles of testing, learning, and adaptation. This is the layer where smaller, more agile services firms often have the edge.

When nobody has all the answers yet, optionality matters. A client working with a nimble agency partner can test promising ideas quickly, learn from real-world results, pivot when needed, and scale what works — all with less cost, less risk, and less delay than a traditional implementation model.

For agencies, that creates a virtuous cycle. More experiments lead to faster learning, which strengthens their cross-client advantage. And for clients, the value is not just execution, but a partner who can help them navigate uncertainty systematically and turn fast feedback loops into forward progress.

I’ll confess a bias: I’ve always loved the creative energy of agencies. The willingness to try something unexpected. To build something bespoke and original for a specific client’s specific problem. That spirit — part artistry, part engineering, part sheer nerve — is the soul of the services business.

AI gives that spirit a much bigger canvas to work on.

The gap between a creative idea and its execution has never been smaller. Concepts that would have taken weeks to prototype can be tested in hours. Campaigns that would have been prohibitively expensive to personalize can now be tailored at a granularity that was previously unimaginable. The constraint used to be “can we build it?” Now it’s “can we imagine it?”

That’s an agency’s home field.

Now, none of this is automatic, of course. The same AI that empowers agencies also empowers clients to do more themselves. Commodity execution will get cheaper, which means the value has to come from genuine expertise, not just extra hands. And plugging into a client’s governed data infrastructure requires trust, access, and technical fluency that not every agency has today.

This also isn’t a zero-sum “services win, SaaS loses” argument. The infrastructure shift is creating tremendous value for platform companies too. SaaS companies that embrace their role as infrastructure, evolve into context-as-a-service platforms, and lean into partner ecosystems can thrive alongside the services firms building on top of them.

But I believe the balance of value creation is shifting.

Agencies and service providers who invest deliberately in AI fluency, build systems to capture and compound their cross-client learnings, and lean into the speed and creativity that are their natural strengths are well positioned to capture it.

The clients navigating this AI transformation won’t find all the answers in a software purchase. They’ll find them in partners who combine deep domain expertise with AI fluency, creative ambition, and the agility to move at the speed this moment demands.

It's springtime for agencies. Time to grow.

Always at your service,

Scott

P.S. Goldcast sponsored today’s newsletter with a free download of the AI Marketing Alliance 2026 Buyer’s Guide. It's a great read, and I bet you’ll discover more than a few surprises (I did).

But I also want to give Goldcast a shout-out as our long-time platform partner for hosting #MartechDay, coming up again on May 5 (Cinco de Martech?). It’s been a fantastic platform for our online events, live and on-demand: fast, flexible, reliable. And their new AI tools for repurposing sessions into blog posts and social media clips turn one live event into a content engine.