For decades, advertising agencies competed on creativity, media leverage, and client relationships. Data played a supporting role, useful, but rarely decisive. That hierarchy is now drastically shifting. In the age of AI, agencies are discovering the same hard truth already reshaping market research and enterprise strategy: data is no longer an input to strategy; data is the strategy.
Artificial intelligence has flattened many historical advantages. Media planning models, audience clustering, creative testing, and even campaign optimization are increasingly automated. What once required large teams of analysts can now be replicated with off-the-shelf models.
Arnaud Frade
Enterprise GTM Lead, APAC
For global advertising conglomerates, this moment is existential. The same AI technologies that promise efficiency also threaten disintermediation. Brands can increasingly run performance media in-house. Retail media networks are absorbing spend once controlled by agencies. Platforms sit closer than ever to first-party signals. In this environment, agencies that merely optimize campaigns risk becoming interchangeable vendors. Agencies that own differentiated data systems become strategic partners.
The Holding Companies Have Seen This Movie Before, But This Time It’s Different
The major agency groups, WPP, Publicis Groupe, Omnicom, Interpublic Group, and Dentsu, have all invested in data for years. But historically, much of that data sat in silos: CRM here, media exposure there, brand tracking somewhere else. AI is forcing consolidation.
Large language models and predictive systems do not thrive on fragmented datasets. They require persistent identity resolution, longitudinal consumer histories, and a steady flow of fresh signals. The result is a strategic reckoning: agencies must move from being integrators of third-party data to owners and orchestrators of proprietary data ecosystems.
Publicis has arguably moved fastest. Its multi-billion-dollar acquisitions of Epsilon and CitrusAd were not about scale alone; they were about control of identity and commerce signals. By 2024, Publicis positioned its CoreAI platform as a unified operating system spanning creative, media, commerce, and personalization, powered by first-party data from over 2.3 billion consumer profiles. The strategic implication is clear: if you control identity, you control outcomes.
WPP, through WPP Open, has taken a different but complementary path, building an AI-native layer that sits across agencies while increasingly integrating client first-party data and retail media feeds. Omnicom’s Omni platform, IPG’s Acxiom-anchored data spine, and Dentsu’s Merkury identity system all reflect the same realization: without owned data, AI becomes a cost reducer, not a value creator.
Media: From Buying Audiences to Owning Signals
Nowhere is the data shift more visible than in media. As third-party cookies disappear and retail media networks rise, agencies are being forced to rethink how audiences are defined and activated.
The most advanced holding companies are no longer optimizing against platform-defined segments alone. They are building persistent audience graphs, anchored in first-party and partner data, that travel across platforms and channels. This allows agencies to move from probabilistic targeting to deterministic, outcome-linked planning.
Measurement and Accountability: Where Data Becomes Power
As CMOs face growing pressure from CFOs, agencies are increasingly judged not on outputs but on business outcomes. AI has accelerated this shift by making multi-touch attribution, marketing mix modeling, and incrementality testing faster and cheaper. But again, the quality of measurement depends on data ownership. AI models trained on incomplete or platform-biased data will confidently produce misleading results. Agencies that control end-to-end datasets, spanning exposure, experience, and transaction, can deliver more credible accountability.
For many organizations, the challenge is not recognizing the value of consumer signals, but generating them consistently and at the speed modern marketing systems require. Traditional research workflows were designed for episodic studies, not for systems that learn continuously. PureSpectrum’s infrastructure was built to make this possible, allowing organizations to generate fresh consumer signals on demand and integrate those insights more directly into the data ecosystems powering AI-driven marketing systems.
The long-term trajectory is becoming clear. The most successful holding companies are no longer positioning themselves as collections of agencies, but as marketing operating systems, where data, AI, creativity, media, and commerce are natively connected. The recent news from WPP, attempting to move away from even recognising itself as a holding company, won’t contradict this new reality.
In this model, data is not a byproduct of campaigns; it is the asset that compounds over time. Every campaign enriches the system. Every interaction improves the model. This creates increasing returns, and switching costs, for clients.
The alternative is stark. Agencies that rely primarily on third-party platforms for data will find themselves squeezed between brands building in-house capabilities and platforms offering self-serve AI tools. Efficiency alone will not save them.
In the AI era, advertising agencies face the same reality confronting market researchers and enterprises more broadly: when intelligence becomes cheap, inputs become priceless. The winners will not be those with the flashiest demos or the most automation, but those with the deepest, most trusted, and most actionable consumer data.
For holding companies, this is no longer a technology question. It is a governance question, an investment question, and ultimately a strategy question. Data ownership determines AI effectiveness. AI effectiveness determines relevance. And relevance determines survival.
The message for agency leaders is clear. Creativity and relationships remain critical, but in the age of AI, data is the strategy, and the agencies that act on this truth will define the next era of advertising.
Publicis Groupe – CoreAI and data strategy
https://digiday.com/media/publicis-groupe-debuts-new-coreai-platform-and-e300-million-ai-investment/
WPP Open and AI-driven marketing transformation
https://www.wpp.com/en/news/2025/10/wpp-unveils-wpp-open-pro
Omnicom Omni platform and data integration
Interpublic Group – Acxiom and data-driven marketing
https://www.acxiom.com/news/ipg-mediabrands-acxiom-and-iris-tv-launch-contextual-ctv-solution/
Dentsu – Merkury identity and data platform