The CX Opportunities in Marketing With Agentic AI

New insights from McKinsey and Company set out how agentic AI is poised to reinvent marketing workflows, automating complex multi-step processes and driving measurable growth. CX leaders should take note as these changes could have far-reaching impacts on the relationship between brands and customers. The same capabilities that promise to transform how marketing departments plan and execute campaigns are also likely to reshape how customers are served, recognised, and retained.

The McKinsey analysis, “representing views from McKinsey’s Growth, Marketing, and Sales Practice”, estimates that agentic AI could eventually power as much as two-thirds of current marketing activities. These are not incremental improvements. The report projects revenue uplifts of 10 to 30 per cent through hyperpersonalised marketing, and campaign creation timelines that compress by a factor of ten to fifteen compared with traditional workflows.

The underlying shift is from what McKinsey calls a "patchwork of disconnected pilots" to genuinely end-to-end agentic systems in which networks of AI agents handle most of the execution, while humans focus on strategy, oversight, and creative judgment. Despite significant experimentation, the report notes that fewer than 10 per cent of CMOs have yet captured meaningful value across end-to-end workflows. The gap between pilot and scale remains wide.

Agentic systems can generate, test, and refine offers, recommendations, and messaging continuously rather than in campaign cycles. Brands that implement these workflows could serve individually tailored experiences to large customer bases without a proportional increase in resource.

Next-best-action decisions that previously required analyst input or batch processing can increasingly be made by agents responding to live behavioural signals. This changes not just the speed of CX but its character, moving from scheduled touchpoints to genuinely adaptive journeys.

As marketing and service data converge within shared agentic infrastructure, the context a customer has already provided in one channel becomes available across others. Agents that know what a customer was offered last week can resolve a service query this week with less friction and greater relevance.

McKinsey highlights the insights function as a priority area for agentic augmentation, including synthesising data, interpreting consumer signals, and translating findings into action. Behaviour patterns that previously sat in dashboards can increasingly be converted into operational decisions in real-time.

AI trust in customer experience is no small concern. McKinsey's survey of 35 CMOs from Fortune 250 companies found that brand and legal governance ranked as the primary worry, ahead of technology underinvestment and data bottlenecks. The report calls for strong governance structures, led from board and CEO level, before scale is attempted.

There are good reasons for that caution. Poorly calibrated agents can make decisions that damage trust in ways that are hard to reverse. Personalisation that tips into intrusiveness creates discomfort rather than loyalty. And agentic systems operating across large organisations risk generating brand inconsistency if content is produced at volume without sufficient oversight. Human judgment, according to the report, remains essential alongside any level of automation.

McKinsey's five-step framework for building agentic workflows offers a practical starting point for CX teams. Begin by auditing fragmented journeys and mapping which repetitive workflows consume the most time. Prioritise first-party data quality, since agent performance is directly dependent on the reliability of the signals they receive. Test agentic AI in contained, low-risk use cases before scaling, and build governance mechanisms before capacity rather than after problems emerge.

The report also cautions against an agents-only mindset. Other tools, including robotic process automation and machine learning, should be considered alongside agentic AI rather than replaced by it.

McKinsey framed this analysis as the reinvention of marketing workflows, but the implications of this go much further. The same infrastructure that enables an AI agent to plan and execute a campaign at ten times the speed also enables a brand to recognise a customer, resolve their issue, and anticipate their next need in ways that were previously impractical at scale.

There are evidently significant opportunities here for enterprise-wide CX transformation, with marketing as the entry point. Brands that move first on agentic AI, and govern it well, could secure advantages in speed, relevance, and customer loyalty that will be difficult for slower movers to close.