Omnicom Adobe Practice says marketers need a new AI operating model

Omnicom’s Adobe Practice has published a whitepaper arguing that many marketing AI efforts stall because companies deploy agents without redesigning workflows, governance and human oversight. The report outlines how brands can split work between AI and people as agentic AI spending comes under pressure to prove measurable business value.

Why it matters: - Marketing teams are spending on AI, but many programs are not producing clear returns. - Omnicom’s Adobe Practice says the missing piece is not more tools, but a redesigned operating model for agentic work. - The guidance comes as Gartner predicts more than 40% of agentic AI projects will be canceled by the end of 2027 because of rising costs, weak business value or poor risk controls.

What happened: - Omnicom’s Adobe Practice released a whitepaper titled “The New Operating Model: How AI Agents and Humans Work Together.” - The report examines why many marketing AI investments have not delivered measurable results. - The whitepaper outlines an operating model that assigns specific work to AI agents and specific work to people. - Omnicom’s Adobe Practice says stalled AI programs often trace back to operating models that were never redesigned for agentic work.

The details: - The practice says agents fail most often when companies put them on unstandardized workflows, feed them siloed data or use them for tasks that did not need autonomy. - The report identifies three common responses to AI in marketing: replacement, resistance and tool pileup. - Replacement means cutting headcount and handing work to agents, which removes human review for off-brand, inaccurate or noncompliant output. - Resistance means delaying adoption until risk feels manageable, while competitors compress cycles and scale personalization. - Tool pileup means adding AI agents to fragmented stacks, which creates more handoffs and more places for work to stall. - Gartner estimates that only 130 of the thousands of vendors marketing agentic AI actually offer genuine agentic capabilities. - Omnicom’s Adobe Practice says the split between machine work and human work should be an explicit design decision. - AI agents should handle governance and compliance pre-checks, routing and workflow orchestration, personalized variant generation, channel and budget optimization, performance analysis and first-draft production. - People should handle brand and campaign strategy, creative concept, final calls on brand fit and accuracy, high-stakes and regulated claims, stakeholder relationships, and ethical and cultural judgment. - The handoffs between people and agents determine whether the model works. - Each human decision must flow back into the system so the next cycle starts with better information. - The company’s framework rests on three layers built in sequence: standardized workflows, agentic orchestration, and governance and human oversight. - Standardized workflows include defined triggers, agent permissions and human review points before deployment. - Agentic orchestration lets agents handle routing, generation, optimization and checks within defined goals and guardrails. - Governance and human oversight include guardrails, audit trails, escalation paths and guardian agents that monitor other agents. - Omnicom’s Adobe Practice says it built its agentic operating model with Adobe. - The solution integrates Omnicom’s Omni platform and agentic framework with Adobe’s content and data stack, anchored in Acxiom’s identity foundation. - The solution is delivered as a governed, managed offering.

Between the lines: - The report argues that AI adoption problems in marketing are often organizational, not technical. - The emphasis on governance and human judgment suggests Omnicom sees agentic AI as augmentation, not a shortcut to full automation. - The focus on employee trust signals that internal resistance can slow adoption even when the tools work. - Workers who fear replacement may avoid helping design new workflows or may route around them. - Omnicom recommends treating change readiness as a dedicated workstream, with leadership clearly explaining how roles will evolve.

What's next: - Marketing, operations and technology leaders can download the full whitepaper on the Omnicom Adobe Practice website. - The full report includes industry use cases, a phased implementation framework and outcome metrics. - Omnicom’s Adobe Practice says the approach is designed to help organizations deploy AI with clearer accountability and better results. - More information is available at the practice website.