A strategic comparison of dentsu, Havas, IPG, Omnicom, Publicis and WPP
As marketing and technology converge, agency platforms are becoming central to how advertising holding companies create value. This report evaluates the flagship platform solutions from six global networks — dentsu, Havas, IPG, Omnicom, Publicis and WPP — to help CMOs assess strategic fit and guide agency leaders in future-proofing their strategic assets. Our comparative analysis draws on independent research as well as market intelligence.
Today, AI-derived “intelligence” is often embedded within proprietary ecosystems, meaning that the value it generates is tightly linked to platform design and governance. By evaluating each platform through a dual lens — client intelligence ownership (critical for CMOs) and platform scalability (key for agencies), our goal is to offer a clearer picture on how this interdependent value chain shapes long-term marketing capabilities.
Intelligence ownership for brands vs agency platform scalability
Amongst agency-built marketing platforms, not all intelligence is created equal — nor equally accessible. The following matrix integrates two essential perspectives: the client’s need for data control and transparency and the agency’s ambition to build scalable, integrated solutions that work across teams.
By cross-examining intelligence ownership from the client’s perspective and platform scalability from the agency’s vantage point, we offer a more nuanced view of platform value. One axis captures how freely and transparently a brand can operate within the system — critical for CMOs seeking to build long-term data capabilities. The other reflects the operational reach and automation potential of each platform — vital for agency leaders investing in future-ready, AI-powered infrastructure.
Together, these two dimensions reveal not just what platforms do, but how and for whom they create value.
X-axis — Intelligence integration: This dimension reflects how intelligence is governed and integrated within each platform:
- Brand-integration: Platforms designed to plug into a brand’s ecosystem—enabling first-party data use, flexible integration, and interoperable identity management
- Agency-centric: Platforms where identity resolution, audience intelligence, and decisioning logic are controlled by the agency
Y-axis — Platform focus: This dimension reflects the platform’s design intent:
- Automation-driven: Platforms built to orchestrate across multiple marketing functions (creative, media, CRM, commerce) using AI/ML to automate workflows, personalize at scale and optimize campaign delivery
- Use case-led: Platforms purpose-built around specific capabilities and intention of offering depth within their core function rather than cross-functional orchestration
Our findings
The approach each agency network takes for their technology and platform investment provides a view on their long-term value proposition and approach to client engagement.
Agency-centric, use case-focused
Merkury is dentsu’s attempt to build a durable first-party identity graph, fused with real-time decision making. The focus is on reducing churn, increasing lifetime value and creating more personalized customer journeys in a privacy-compliant way. It is positioned as a direct client value driver, not just an optimizer for media.
Implication: Strong fit for brands seeking outcome-driven media tied closely to business KPIs in the US.
Moderate brand integration, balanced focus of automation and use case depth
Converged is intended to be a platform to integrate its creative, media and health businesses into a single experience offering. Though relatively early, it presents Havas’ forward-looking model centered on human experience and cross-functional data alignment
Implication: Fairly new technical offering compared to industry peers and presents a vision aligned with brands looking for meaningful differentiation through customer experience and more flexible and tailored agency relationships.
Balanced focus of brand integration and use-case focus
Operating on its core platform Interact, Kinesso offers a modular, privacy-first approach to audience intelligence. Built to integrate with Acxiom’s extensive data infrastructure, it supports customizable configurations across media and CRM, positioning itself as a post-cookie solution for brands seeking greater control and transparency.
Implication: Best suited to brands seeking transparency and control over data and activation. Its privacy-centric stance will appeal to regulated industries and privacy-conscious CMOs.
Agency centric with strong focus on automation
Omni 3.0 is designed as an open operating system, enabling rapid adoption of generative AI models and seamless integration with a range of technology partners. Tools like Omni Assist make it a powerful orchestration engine across media and creative, with robust AI capabilities that streamline workflows and boost operational efficiency. However, while Omni supports interoperability at the execution layer, core data assets (such as Flywheel, Omnicom’s commerce intelligence solution) and key identity components remain agency-controlled, which can limit clients’ ownership of the underlying intelligence.
Implication: A strong orchestration solution for brands needing speed, personalization and campaign precision.
Strong brand integration and platform automation
Publicis positions CoreAI as the “intelligence layer” across the Groupe. It is vertically integrated through Epsilon (identity/data), Sapient (experience/consulting), and media/creative arms, aiming to control the full marketing stack. With its recent acquisition of Lotame, Publicis again makes a clear statement that the future of advertising lies in data and platform ownership. Lotame coupled with Epsilon gives clients the ability to integrate their own first-party data with this massive identity graph, giving them greater control and flexibility in how their data is used for marketing, measurement and activation across regions and channels.
Implication: Full-stack solution well-suited for global brands seeking one partner for integrated services.
Strong brand integration and platform automation
Open is marketed as a client-operable operating system rather than a walled garden. It integrates with clients’ tech stacks, built on AWS and Azure and powered by partnerships like IBM’s WatsonX. Open’s open architecture and deep integration capabilities allow an end-to-end orchestration across the entire customer journey. WPP can bring insights, creative, media and measurement best-in-class tools, data and AI all within a single, flexible platform.
Implication: Ideal for brands with mature data infrastructure and a desire to scale intelligence across the stack without sacrificing agility or ownership
Platform differentiators: architecture, data & AI capabilities
Platform architecture: Open, closed, modular
- Open: Can plug into external tools, client systems, or third-party data providers via APIs (e.g. WPP Open, IPG Kinesso)
- Closed: Operates within a walled garden, optimized for agency workflows (e.g. CoreAI, Omni)
- Modular: Built in blocks—brands can pick and choose which pieces to use
Why it matters for brands: Open/modular platforms allow more flexibility and integration with the brand’s own tech stack—ideal for brands with existing data literacy or marketing ops expertise. Closed platforms offer speed, consistency and turnkey outcomes, but often at the cost of vendor lock-in and data portability.
Data: Borrowed, owned
- Owned data: Data that comes directly from the brand’s touch points (CRM, website, transactions)
- Borrowed data: Data obtained via third-party sources or shared IDs (identity graphs, partnerships)
Why it matters for brands: Platforms built on deterministic, owned data (like Epsilon’s Core ID or Acxiom’s Real ID) provide accuracy and better long-term value, especially for personalisation and retention. Borrowed or probabilistic data can be useful for scale, but offers less control and faces increasing regulatory risk (e.g. cookie deprecation, privacy laws). Ownership is key: when the agency controls the primary identity layer, the brand’s ability to migrate or integrate data independently is limited.
AI capabilities: Embedded vs integrated
AI in agency platforms isn't one-size-fits-all. The real differentiator is how the AI models are trained, what data they ingest and whether they exist to improve internal agency workflows or deliver client-facing intelligence.
- Embedded AI typically means AI is built deeply into a platform, often to enhance its core functionality or automate workflows. It’s “baked in” and often automates internal workflows and enhances operational efficiency
- Integrated AI refers to AI that is connected or layered onto an existing system, often to provide new capabilities or insights. It can be more modular, allowing for the addition of external AI models (like GPT or Claude) to augment processes or deliver client-facing intelligence
The real competitive edge comes from platforms that can combine proprietary, high-quality data with AI to deliver unique, client-facing intelligence. Proprietary data, such as Acxiom (IPG) or InfoSum (WPP), allows agency platforms to fine-tune models for accurate targeting and personalization in a privacy-first way. This then creates a “moat” that generic, off-the-shelf AI cannot match. This is why most holding companies are investing heavily in owning and integrating unique data assets into their platforms.
Why it matters for brands: Brands should look past the AI hype and scrutinize whether a platform’s AI is truly making their marketing smarter through proprietary data-driven intelligence, or simply making agency operations more efficient. As computing becomes increasingly expensive, how brands extract value from their agency partner’s AI & data capabilities is a critical aspect of evaluation.
Depth vs Scale
In today's marketing infrastructure, scale typically refers to how broadly a platform can operate - its reach across channels, markets, devices and datasets. It's about volume, consistency and cost-efficiency. On the other hand, depth (or scope) is about how richly a platform can operate—its ability to handle complexity, context, brand nuance and customer-specific experiences.
- Scale equates efficiency and cost optimization: It automates and standardizes, making large campaigns faster and cheaper to deploy
- Depth makes the money work harder: It increases precision, resonance and lifetime value through specific capabilities solving specific use cases
In an ideal world, platforms would offer both strategic depth and effortless scalability. But in reality, most agency platforms lean toward one or the other. This matters more than ever, because brands now engage with consumers across a complex constellation of touchpoints — from connected TV and retail media to brand-owned CRM systems, apps, social platforms and even influencers.
Platforms with depth tend to focus on strategic integration within a brand’s own ecosystem — emphasizing specificity, interoperability, modular design and customization. They may scale less efficiently across global footprints but offer greater ownership and co-creation potential for in-house teams. Scope-specific platforms, such as Merkury, allow brands to tune experiences to specific channels and contexts. Scaled platforms, by contrast, are optimized to deliver consistent messaging across paid media channels — powerful for reach and efficiency, but potentially limiting in emotional nuance and contextual sensitivity.
Brands don’t need to be everywhere — they need to matter. And that requires deep insight into moments, behaviors and emotional context.
For global brands, especially in performance-driven verticals, scale is non-negotiable. But for niche, localized or high-value segments, scale without precision leads to waste. Worse, relying on scaled datasets trained on U.S. consumer behaviour (the biggest media market in the world) can create cultural misfires in markets like APAC, LATAM or Europe.
Strategic implications for brands: roadmap for decision making
The platform that brands select through their agency must align with their long-term operating model for how the brand engages with consumers, how it uses its data, and how it scales intelligence across every touchpoint.
Who owns the intelligence?
Modern marketing is built on data, but not all platforms treat it equally. Some give you full transparency and integration into your stack. Others offer outputs but hold the underlying intelligence.
Ask yourself:
- Can we plug in our own first-party data?
- Do we get visibility into customer journeys, or just performance reports?
- Can we extract, reuse or port data when needed?
- Is the platform optimizing for us, or optimizing agency workflow?
Is AI producing value beyond operational efficiency?
All platforms now claim AI capabilities, but their purpose, training data and control structures vary dramatically.
- Client-based AI leverages owned, deterministic data, as well as, brand guidelines to power CRM, media and commerce. This enables platforms to deliver branded, personalized experiences and smarter targeting at scale. This directly impacts customer outcomes
- Workflow AI is built for internal efficiency that helps planners write briefs and build audiences faster. While this helps clients get more for less which increases business outcomes but may not translate to better customer experience
Proprietary data emerges as a true differentiation that enables AI to create very personalized, on brand experiences that off-the-shelf models cannot. This “beyond efficiency” mindset is key to building stronger, more meaningful customer relationships.
Ask yourself:
- Is the platform’s AI improving our personalization, performance and planning?
- Is it trained on our data, or just third-party foundation models? Remember AI trained on your data is your asset
- Can we collaborate with the AI layer, or is it locked inside the agency?
- Are we willing to share proprietary data on our customers and IP with our agency partner?
Do you need scale or scope to win?
Some platforms are optimized for enterprise-wide standardization (scale). Others focus on specific use cases like loyalty, retail media or CRM (depth).
Ask yourself:
- Do we need scale across regions and teams, or depth in specific use cases like loyalty, retail media or CRM with highly contextualized, localized execution?
- Will this platform integrate with our tech stack?
- Will it make our customer experience more seamless without adding operational complexity?
- What’s the cost of the agency / vendor switch, especially with potential data and ‘intelligence’ transport?
Will this platform increase our own marketing capabilities and digital maturity?
A platform that only delivers outputs may speed up campaigns, but it won’t necessarily build your marketing’s long-term digital, data and innovation maturity.
Ask yourself:
- Does it teach us how to operate in a digital-first way — or does it do the thinking for us?
- Can we use it to train our own models, optimize our own journeys, and measure our own outcomes?
- Will our team walk away from each campaign with new knowledge, new capabilities or new IP? Do we prioritize outcomes or insights?
- Is this platform helping us build internal maturity in data, AI, customer experience and agile operations?
- If the agency were to disappear tomorrow — what would we retain?
Ready to learn more? Check out the full research analysis including specific evaluations of each agency platform solution.
