Using Proxy AI to Revitalize the Marketing Workflow (English) 2026

workflows with agentic AI For marketers, AI-enabled workflows will fuel new levels of growth, speed,and efficiency. What is your activation plan? This article is a collaborative effort by Dianne Esber, Eli Stein, Julien Boudet, and Kelsey Robinson, with NilayShah, representing views from McKinsey’s Growth, Marketing, and Sales Practice. The future of marketingwill be defined by how well organizations operate in an AI-mediatedworld. Consumers are discovering, evaluating, and purchasing through increasingly intelligentsystems; attention is fragmented across proliferating platforms; and expectations for relevance,personalization, and immediacy are rising at once. Marketing is no longer confined to campaignsand channels—it is becoming a real-time growth engine that integrates insights, content, That kind of execution is no simple task—and marketing organizations understand this betterthan most. Marketers, after all, have been among the earliest adopters of gen AI, piloting usecases from copy generation to image creation. Many tools have gained traction, yet becausethey typically solve isolated tasks, the result has been a patchwork of disconnected pilots andsystems that increase activity (for example, more early-concept images produced) while Agentic AI—systems built on foundation models capable of acting and executing multistepprocesses—has the potential to address this problem because it offers the opportunity fororganizations to fundamentally transform the way work gets done. Rather than relying onpractitioners using isolated tools to boost individual productivity and effectiveness,organizations can create hybrid human–agentic workforces—in which people design andoversee networks of AI agents that handle most of the execution. In this model, one marketing Realizing this potential value is only possible through the reimagining and rebuilding ofworkflows around agentic AI. This is no simple task, which helps explain why companies so farhave struggled to extract significant value from AI agents. Organizations that fail to do the hard While we are still in the early days of agentic AI, a recipe for how to reimagine and rebuildmarketing workflows is emerging. This article will examine the five-step process for creating an The value of agentic AI in marketing We estimate that agentic AI will come to power as much as two-thirds of current marketingactivities, enabling tasks such as automated content generation, synthetic audience testing, and Exhibit1 Ultimately, an agentic workforce has the potential to transform marketing operations in three Powering topline growth. Organizations that are implementing agentic workflows in marketingcan expect to see 10 to 30 percent revenue growth from hyperpersonalized marketing, always-on, AI-enabled campaigns with improved cross-functional collaboration across teams Enabling speed.We estimate that agentic systems will accelerate the creation and execution ofmarketing campaigns by ten to 15 times, by speeding up both the brainstorming and vetting of Fueling working spend and growth.Powering more work with AI agents will allow resourcespreviously spent on processes and operations to be reallocated toward directly reachingconsumers. The result: humans focusing on the more important tasks and higher ROI from data- These gains, of course, are by no means certain. They will only be realized by reimagining theway marketing work is accomplished. Below, we explain how leading organizations are doing Creating an agentic marketing workflow Designing an agentic AI solution generally requires a five-step process—from identifying thetasks that can be accomplished with agents to rethinking human roles for proper oversight(Exhibit 2). As they navigate this process, leaders must be aware of several factors that add toexecution complexity. Some agentic solutions, for example, can be applied to similar tasks Exhibit2 Step 1: Create a detailed taxonomy of key marketing activities Creating tomorrow’s agent-driven workflow, of course, cannot be done without first developinga granular understanding of the way work gets done today. An important first step in thatprocess is to break down priority workflows into the full chain of key activities involved. Thismapping must include the underlying systems—customer relationship management, content This is how many companies across industries have begun. Take, for example, one leadingconsumer brand that sought to redesign the process of creative ideation and production.Historically, this was an often complex undertaking that could take months of effort, withnumerous stakeholders, both internal staffers and outside agencies, engaged in iterative cyclesof feedback and rework. To determine how AI agents might help, the organization first created acomprehensive list of activities involved in the process, encompassing ideation, conceptcreation and testing, content production, content versioning, content optimization, and agency This taxonomy