Production Has Five Moats

WPP just announced "the world's largest and most advanced content production powerhouse." Nearly 10,000 people across 40 cities. Two numbers that would concern me as a client or investor (and certainly as someone among those 10,000).

Three holding companies now have branded production platforms. Omnicom Production launched in 2025. Publicis Production was already in place. And now WPP Production (effectively for now the rebranded Hogarth with Richard Glasson at the helm) completing the set. Each claims scale. Each claims innovation. Each claims AI-powered transformation.

However, production is a commodity. And when something becomes a commodity, competing on size is like two airlines racing to buy the biggest fleet. Passengers don't care how many planes you own. They care whether their flight is on time, their luggage arrives, and the food is edible. Scale is a cost, not a differentiator.

The ability to produce content is table stakes. Every one of these platforms can do it. The question isn't who can produce. It's who can produce in a way that clients can't easily replicate themselves, that platforms can't disintermediate, and that actually gets better with scale rather than simply bigger.

That requires five specific moats.

Before the moats, let's agree why production is becoming a commodity in the first place.

To be clear, I'm talking about downstream production: the execution, rendering, and distribution of content after creative master works and strategic decisions have been made. That upstream work (creative strategy, human imagination, client vision) is absolutely a moat, and one of the most critical. This piece focuses on what happens after those decisions are made.

The economics are brutal and accelerating. GenAI has collapsed the marginal cost of producing content. Platforms like Meta and Google are building production tools that clients can use directly. Adobe and Canva are working to skip the agency layer entirely. Clients are in-housing adaptation and personalisation, an adjacent step after in-housing biddable media as they drink from the same data well.

The result: price deflation on a per-output basis. Every output costs less to produce than it did twelve months ago. Clients know this, so they pay less.

But here's the temporary reprieve that masks the structural problem. Diminishing returns to targeted content haven't fully kicked in yet. For years, agencies targeted media to individuals. Now they finally have the technology to target specially rendered content to individuals. That targeted content is more effective, and as long as it is, agencies can offset lower output costs with more volume. Therefore clients are for now opting to produce more. Charge less per unit. Maintain revenue.

This works until it doesn't. Volume has a law of large numbers limit. Costs don't have a lower limit. The economics of production are heading toward a floor that doesn't exist.

Which is precisely why these platforms matter. But not for the reasons being celebrated.

The race isn't about capacity but about about who builds the deepest defences against the forces trying to commoditise them entirely. There are five moats. All five matter.

Moat 1: Share of internal production

Scale effects only materialise if you actually capture the production work that already exists within your holding company. This sounds obvious. It isn't.

When production has been nested as revenue within agency P&Ls, consolidating it into a central platform means those agencies are donating revenue to a new entity. In a culture where each network brand is measured on its own financial performance, this is a painful internal reorganisation not done by a press release.

WPP's structure makes this particularly difficult. The P&L is still fundamentally aligned to network brands. Agencies that have been generating production revenue for years are now expected to hand it over to WPP Production. Under financial pressure. In a culture that is, by most accounts, struggling with morale after years of restructuring.

I'll give WPP this: they are quite good at announcing things. The harder work of remapping teams, revenue, and incentives begins on February 23 when the transition formally takes effect. They do not have the culture to make this successful at the speed that it needs to be successful.

Moat 2: Technology that improves quality

Every production platform is using the same foundation AI tools. The same generative models, the same image generators, the same video synthesis capabilities. The moat isn't the tools. It's the workflows that stitch them together.

The best production output comes from a "best of breed" approach to GenAI point solutions, assembled into modular workflows that evolve as new capabilities emerge. Vendor announcements arrive weekly. The platforms that win will be the ones that can integrate new capabilities fastest without breaking existing workflows.

This is an engineering problem, not a creative one. And it requires a culture of continuous technical adaptation, not quarterly reporting cycles.

Moat 3: Data that improves quality

Here's where the conversation gets uncomfortable for two of the three platforms.

Master content is still briefed by creative teams working with strategy as it should be. But everything that follows such as adaptations, personalised variants, atomised assets for different audiences and channels is briefed by data. Specifically, by the same identity and media data that drives programmatic decisions.

This means production needs to be closer to media and data infrastructure. Further from creatives and their ideas.

Publicis integrated Epsilon as CoreID which is the foundational identity layer powering the entire holding company. Production decisions at Publicis are already data-driven at a scale the others can't match today.

Omnicom now has Acxiom. But Acxiom operates as a separate agency with its own P&L, selling services as a product line. That's cross-selling. Not integration. The structural difference between these two approaches is the difference between a production platform that gets genuinely better with data and one that offers data as an add-on you have to purchase separately.

WPP's personalisation data position is the weakest of the three. No comparable first-party data asset. WPP Open promises AI-powered workflows, but without the identity infrastructure underneath, those workflows are producing content without the targeting precision that makes production genuinely valuable at scale. Some will argue (watch the comments below) that federated third-party approach is sufficient, but this is not a defence against platform disintermediation, as they have the highest definition first-party data of anyone.

Moat 4: Technology that removes friction

Speed is a strategy. Speed of adoption within the holding company. Speed of process from brief to output. Speed of governance and approval. And critically: speed of distribution.

The goal is to produce and distribute at the speed of data. Removing friction at every point in that chain is a compounding advantage. Every day of delay between content creation and distribution is a day of diminishing relevance.

This is where WPP Open's single platform approach has genuine potential. If the technology works as described then operating on a unified system could meaningfully reduce internal friction. The question is whether the organisational culture allows the speed the technology enables.

Moat 5: Technology that protects brands

When you're producing more content, faster, using automated tools and automated distribution, brand safety risks multiply proportionally. Copyright infringement from GenAI-generated works. Inconsistent brand treatment across thousands of automated outputs. Cultural sensitivity failures in localised content.

This is the moat nobody wants to discuss because it admits the risk inherent in the entire production acceleration strategy. But it's real. And clients are increasingly aware of it.

The platform that solves brand protection at scale and not through manual review, but through automated compliance that keeps pace with automated production, will command significant premium pricing.

Publicis leads on all five moats. The Power of One organisational restructure made internal production consolidation structurally easier than it will be for WPP. The Epsilon integration gives them data-driven production at a scale the others haven't matched. Their engineering culture (where roughly one in six employees is an engineer) enables the continuous technical adaptation that Moats 2 and 4 require.

Omnicom has made meaningful progress on technology and is now integrating the Acxiom asset. But until Acxiom operates as infrastructure rather than a product line, their data moat remains structural rather than functional.

WPP has the scale announcement. Nearly 10,000 people is impressive on paper. Smart readers see that number for what it is: a massive restructuring cost. Which makes it an odd number to lead with. But the internal consolidation challenge, the data gap, and a culture shaped by quarterly pressure make converting that scale into competitive advantage the hardest task of the three.

I've only covered the 3 main holdcos today, and needless to say there are strong focused competitors for them, such as Storyteq and Pencil/Oliver.

Back to that airline metaphor. The biggest fleet doesn't win passengers. The airline that gets you there on time, treats you well, and doesn't lose your luggage does. That is what these five moats are about. An inspiring message, in the right place, for the right person at the right time at the right cost.

Brandflow is written by Justin Billingsley, who has spent his career on all three sides of the industry's table: senior client, global agency leader, technology founder. First published 2 February 2026 in the Brandflow newsletter on LinkedIn.