The Future of Agencies in the Age of AI

Last week I spoke at the Google Partner Summit in Dublin about managing agency change in the era of AI, and then moderated a panel of exceptional agency leaders including Pats McDonald, Christina Lemieux, and Liam Wade. In my talk I spoke about the need to combine optimisation and efficiency focused efforts with transformational thinking to reinvent processes, workflows and structures. As part of the talk I thought it would be interesting to do some AI-assisted scenario planning on the future of agencies (in much the same way as I did recently for the future of strategists and planners). What came out of that work was some divergent challenges and pressures but also some real opportunities. It was a useful way to show the different potential choices that are open to agencies, and the risks and benefits of different pathways into the future.

I couldn’t include any more than a high level overview of the scenario planning outputs in my talk so I promised that I would write it up here. Even if you don’t work in advertising I think this could well be of interest. Few industries sit closer to the intersection of creativity and technology than advertising and it’s clear that the sector is already being profoundly reshaped. AI is collapsing production cycles, democratising powerful tools for creativity, and reconfiguring the boundaries between clients, agencies, and platforms. But it also holds the potential to enable huge efficiences, gains in effectiveness, and to augment creativity in totally new ways. The different paths and choices open to agencies now will determine not only whether they will survive and thrive, but also what kind of creative organisation they want to become.

Let’s begin by looking at four fundamental drivers and dynamics of change that are likely to be key to determining the future.

Technological drivers

• AI integration across creative workflows: The degree of integration from augmentation (where AI is acting more as a co-pilot in ideation and production) through to high automation where agentic systems generate, test, and optimise campaigns end-to-end.
• Agentic orchestration: The growing sophistication of interconnected AI agents that are capable of managing projects, coordinating production, and adapting creative dynamically.
• Platform dependency: Whether agencies build their own AI ecosystems or become reliant on third-party systems, for example from OpenAI, Adobe, or Google.
• Data infrastructure and proprietary intelligence: The extent to which agencies can source and build differentiated, high-quality data pipelines to feed creativity, insight, and targeting.

Human and organisational factors

• Creative identity and adaptation: How agency talent redefines its craft, and the balance of human originality and taste with orchestration and system design.
• Leadership mindset and trust: Whether leaders see AI as a threat to creative culture or as an amplifier of it.
• Capability transformation: The potential for new hybrid roles who design with, not against, machines.
• Culture and adoption: How fast agencies can shift from experimentation at the edges to systemic integration across teams.

Market and economic context

• Client in-housing and self-service tools: The degree to which clients internalise AI capability and reduce reliance on external agencies.
• Evolving business models: From retainers and billable hours to value-based or outcome-driven models.
• Industry consolidation: The potential for platform–agency alliances or the rise of AI-native creative boutiques.
• Procurement and performance pressure: The push for cost efficiency versus the pull for breakthrough creative impact.

Societal and cultural shifts

• Ethics, authenticity, and provenance: How audiences and clients respond to AI content and ‘synthetic creativity’. Whether human-made retains cultural premium value.
• Cultural fatigue and trust: Whether the abundance of AI-generated content diminishes distinctiveness or drives renewed demand for meaning and craft.
• Regulation and accountability: Emerging rules around data, copyright, and transparency that could define how agencies work with generative systems.
• Human creativity as signal: The extent to which ‘made by humans’ becomes a badge of distinction in a world of algorithmic output.

Across these dynamics, several underlying themes stand out: the move from production to orchestration, the role of trust and ethics, differentiation through data and the risk of sameness at scale, shifting client-agency relationships and power dynamics, the degree to which AI is seen primarily as an efficiency-driver or a creativity-amplifier.

So, to articulate how these forces could intersect and unfold I’ve mapped out three dynamics critical to shaping possible futures for agencies. To avoid a surfeit of matrices I’ve summarised these and then brought them all together in two 2 x 2s at the end which show four potential futures for agencies and four paths for AI transformation.

Framework 1: Creative identity

This model looks at how the centre of gravity in creativity will shift with AI. One axis positions human-led craft (empathy, taste, storytelling, cultural intuition) against AI-orchestrated creation (speed, scale, and data-driven imagination). This is juxtaposed against high or low levels of client-agency trust and regulation to shape four ‘snapshot scenarios’:

1. Human-crafted trust. Agencies double down on human craftsmanship, authenticity, and creative ethics. Clients value provenance. AI is used sparingly as a behind-the-scenes accelerator. High trust, slow experimentation.
2. Synthetic renaissance. A golden age of AI-powered creativity where agencies fuse art and algorithm. Clients embrace synthetic storytelling, hyper-personalisation, and generative experimentation. Distinction comes from how well humans orchestrate AI systems.
3. Algorithmic anarchy. Low regulation and low trust lead to creative overproduction and content fatigue. Agencies flood media channels with auto-generated work; clients and audiences become desensitised. Speed beats craft until trust collapses.
4. Ethical optimisation. Agencies lead the way in responsible orchestration with certified ethical frameworks and transparent AI pipelines become selling points. Differentiation lies in combining regulatory fluency with creative agility.

This highlights how the creative process and client-agency relationship may evolve as AI reshapes the meaning and implementation of originality and creativity.

Framework 2: Power and structure

This lens explores who will control the creative value chain in an AI-first industry.
On one axis, client and platform control versus agency orchestration. On the other, fragmentation versus integration of agentic ecosystems. The resulting four mini-scenarios for this one are:

1. DIY marketing. Clients use AI-first marketing stacks and internal teams to create content on demand. Agencies shrink into specialist consultancies. Speed and control win over creativity.
2. The Superagency. Agencies evolve into networked platforms orchestrating multiple AI agents, creative systems, and data flows. Human strategists focus on design, governance, and innovation across ecosystems.
3. Tool chaos. No clear standards emerge. Agencies and clients battle interoperability issues, tool fatigue, and workflow fragmentation. Creativity slows under the weight of complexity.
4. Hybrid studios. Agencies and clients co-create within shared AI workspaces. New partnership models emerge around shared data and agentic creativity. Trust and transparency become the new currency.

This is really about how the ownership of data, tools, and agentic systems could determine who captures the most value in the future (clients, platforms, or agencies themselves).

Framework 3: Value and strategy

The third model looks at whether agencies use AI primarily for efficiency or for reinvention, highlighting the difference between using AI to optimise existing processes, or to reimagine the agency operating model entirely. So one axis is efficiency-led versus creativity-led transformation, and the other is tactical AI adoption (tools as add-ons) against strategic reinvention (an AI-native operating model). Our four mini scenarios from this are:

1. Automation trap: Agencies automate production and reporting but fail to reinvent their strategic value. Efficiency gains are real but short-lived, and differentiation evaporates.
2. The Reinventors: Agencies rebuild from the ground up around AI-native principles including agentic workflows, adaptive teams, and new value propositions. Creativity and efficiency coexist symbiotically.
3. Legacy operators: Minimal adoption beyond tactical tools. Agencies cling to legacy structures, talent drains to more adaptive competitors.
4. Creative ‘superfusion’: Generative AI becomes a creative co-pilot across strategy, concepting, and production. Agencies evolve into continuous invention systems where human taste directs infinite AI possibility.

This framework asks the central strategic question of whether AI will make agencies faster, or fundamentally different?

Four potential futures for agencies in the era of AI

Bringing these dimensions together we can set out four overarching future scenarios for agencies in the era of AI, in a way that considers the intersection of creativity, control, and value:

These four futures illustrate the critical choices now facing agency leaders, which is whether to compete on automation or on orchestration, and whether to root creativity in technology or in human imagination. The central question is how to build an organisation that can harness AI’s scale and speed without losing the cultural depth and meaning that make creativity valuable in the first place.

Four paths for transformation

We can also set out four possible paths that agencies may take to get there. How they might approach their own AI transformation.

This second matrix highlights the internal choices that will define each agency’s way forwards. Ultimately, the future of agencies will likely hinge less on technology itself and more on how leaders choose to respond to it. AI can be both a mirror and a multiplier. The choices of how it is applied can reflect both how an agency sees itself and also amplify its strengths or weaknesses.

Agencies can rush to efficiency and optimise what’s already there, or see it as a catalyst for reinvention. They can pursue efficiency alone and risk commoditisation, or use AI to build new kinds of intelligence and imagination into their work. They can operate as suppliers in someone else’s system, or as orchestrators of systems that blend human judgment with machine capability.

The next few years are definitely going to be interesting.

A version of this post appeared on my weekly Substack of AI and digital trends, and transformation insights. To join our community of over ten thousand subscribers you can sign up to that here.

To get posts like this delivered straight to your inbox, drop your email into the box below.