HBR’s premise is blunt: AI sped up engineering first, and marketing is now stuck doing quarterly-era coordination in a continuous-release world.
Citing Anthropic research, the piece notes that software engineering accounts for “nearly 50% of all agentic activity.” That helps explain why product teams are shipping continuously, and why marketing — with its cross-functional handoffs, siloed systems, and sequential approvals — suddenly looks like the slowest dependency on the critical path.
The article concedes marketing has already seen real AI gains in copy generation, image creation, and personalization. But it argues those are “localized” improvements. Faster drafts do not equal faster launches when the operating model is still built like a relay race with too many batons.
The proposed fix is an “agentic marketing organization,” which is less a new tool stack than a new workflow architecture for human-agent collaboration. At the center is a “brand code”: a machine-readable knowledge base that codifies brand strategy, product experience, customer insights, and business rules into structured formats like taxonomies, prompt templates, decision trees, and tagged datasets.
In theory, both humans and agents use that shared base to make decisions and generate work. Over time, performance data feeds back into the system to refine messaging, audience definitions, and decision logic. It is also positioned as a hedge against the familiar problem of institutional knowledge walking out the door with someone who knew where all the bodies, decks, and naming conventions were buried.
On top of that foundation, HBR lays out a layered system: specialized agents for execution, an orchestration layer to route work and manage dependencies, and an interface layer embedded in tools like Slack, WhatsApp, or Teams. The marketer’s role becomes less “open eleven tabs and chase approvals” and more “set intent, review outputs, and decide when the machine needs adult supervision.”
HBR points to early implementations at companies such as HubSpot and AWS, with some large numbers attached: marketing materials adapted “up to 98 times faster,” unit costs reduced by 80%, and click-through rates increased up to 17 times. The article does not unpack the methodologies or baselines behind those figures. It also cites BCG research finding that organizations embedding agentic AI into marketing workflows can achieve up to a threefold increase in ROI, campaign speed, and content volume.
The tension the piece does not hide: this is not just a capability shift. It is an identity shift. Marketers who built careers on doing the work are being asked to become directors of systems — judging outputs, tuning workflows, and deciding what good looks like instead of polishing every asset themselves.
Read more at Harvard Business Review.