The vocabulary.

Marketing executed end-to-end by a multi-agent AI system, without a human in the critical path.

Autonomous marketing is a category in which AI systems generate hypotheses, produce assets, ship campaigns, read responses, and update their own model of the brand — operating as a system, not as tools used by a human marketer. Coined and developed by Chandan Kumar. Distinct from AI-assisted marketing, which keeps humans as the initiator of every task.

Related: Multi-agent orchestration · Brand memory · Recursive learning

A structured, queryable model of a brand's voice, audience, performance history, and failure modes.

Brand memory is the substrate that lets an autonomous marketing system compound learning. Every piece of content shipped, every response read, every experiment run becomes durable context for the next decision. Unlike a prompt or a vector dump, brand memory is structured — queryable, updateable, and designed to outlive any single campaign or model.

Related: Autonomous marketing · Recursive learning

An AI system's ability to change its own decision-policy based on performance data.

Recursive learning goes beyond collecting metrics — it is the property of a system that updates how it decides based on what it has observed. In autonomous marketing, recursive learning is what makes a running system a moat: a brand whose system has been learning for eighteen months is uncatchable by a brand whose system just turned on.

Related: Autonomous marketing · Brand memory

The current moment in autonomous marketing — the beginning of the category.

Year 0 is the framing Chandan Kumar uses for the present phase of AI in marketing. The components exist (capable LLMs, tool use, memory substrates, evaluation loops) but the integration — the playbook — has not yet been written. Year 0 is the strategic window to either write the playbook or have it written for you.

Related: Autonomous marketing

Acquisition, retention, and monetization as three layers of one system.

A framework that rejects the typical marketing org-chart split into separate acquisition, retention, and monetization teams. The stack is read bottom-up during optimization (acquire right, retain, then monetize) and top-down during diagnosis (check monetization and retention before blaming acquisition). Four common failure modes: Leaky bucket, Locked vault, Undervalued asset, and Disconnected stack.

Full framework →

Related: Stage-Gate Marketing Model

The coordination of specialized AI agents — creative, distribution, listening, experimentation — by a planner.

The architectural pattern behind autonomous marketing. A single model cannot hold a whole brand's marketing context and execute every function. Multi-agent orchestration distributes work across specialists with defined handoff protocols, making the system extensible and debuggable.

Related: Autonomous marketing · Brand memory

A marketer who operates across the full revenue lifecycle, not just one channel or layer.

Most people in the AI-plus-marketing space are either pure AI technologists who don't understand marketing, or marketers who are dabbling in AI tools. A full-stack marketer has lived inside acquisition funnels, retention programs, monetization experiments, and revenue growth strategies — at both startups and enterprise scale. The 360-degree view is rare and is the foundation of the perspective Chandan Kumar writes from.

Related: The Revenue Lifecycle Stack

Why the marketing playbook that works at ₹1Cr breaks at ₹10Cr and again at ₹100Cr.

A stage-by-stage framework for what to scale and what to kill at different revenue milestones. Founder-led growth runs out at ~₹5Cr; early channels turn into efficient markets; retention becomes the denominator; the first marketing hire often needs to be redesigned. See also the essay "What breaks at ₹10 crore."

Related: The Revenue Lifecycle Stack