Unpacking Agentic Commerce
For three decades, commerce has centered on human buyers navigating browsers, apps, and IoT devices. Humans absorb ambiguity — tolerating unclear checkout processes, mysterious promotions, shifting shipping estimates, and surprise totals. If the brand is strong enough, they might still purchase despite friction.
Agentic commerce marks a fundamental shift: acknowledging that buyers aren't always human. Automated agents operate differently from people. They lack emotional attachment to brands and won't retry after failures. Instead, agents follow intent within constraints and systematically avoid unreliable paths.
The Core Challenge: Architecture Must Change
The real pressure isn't on shopping experiences — it's on the systems behind them. This affects architecture, retail operations, and manufacturing in sequence, revealing a unified challenge: making promises legible to machines.
Current systems expose internal plumbing as products. While microservices enable decomposition, they create accidental interfaces that humans navigate through documentation and support. Agents don't adapt; they select alternatives. When the buyer is automated, the cost of brittleness is no longer a frustrated developer — it's demand quietly routing elsewhere.
The missing element is not additional APIs but a capability surface — stable contracts describing what systems can accomplish, under what conditions, with what guarantees. Capabilities differ from services: services describe how you build; capabilities articulate what you promise.
Retail's Uncomfortable Reality
Retail traditionally optimizes funnels designed for human psychology: attract, persuade, capture, convert. This framework collapses when agents handle purchasing. Agents arrive with clear intent, not susceptibility to persuasion.
Success requires trustworthiness over experience appeal. Critical shifts include:
- Inventory truth matters more than marketing presentation
- Delivery accuracy supersedes persuasive copywriting
- Explicit policies replace ambiguous guidelines
- Deterministic returns processes eliminate judgment calls
- Audit trails become outward-facing requirements, not internal compliance footnotes
Retailers must become reliable counterparties to automation — through capabilities like quote, reserve, promise, order, modify, and return that operate predictably or explain themselves transparently.
Manufacturing's Second Chance
Manufacturers historically relied on intermediaries for discovery, quoting, configuration, service, and distribution — not as gatekeepers but as friction absorbers. Agents automate friction handling, making discovery, procurement, and comparison programmable.
This weakens economic arguments for intermediaries controlling access, though it doesn't mandate dramatic "going direct" strategies. Instead, manufacturers gain opportunities to participate directly in demand flows without human translators.
The requirement is precision. Agents filter out incomplete semantics and document-dependent specifications. In this environment, "brand" becomes measurable behavior: lead time accuracy, defect rates, warranty friction, documentation completeness, and promise-reality alignment.
This is the most inconvenient truth of agentic commerce: trust becomes measurable, and measurement changes power.
Channel Evolution
Intermediaries survive through operational value — local fulfillment, services, installation, financing, support. Those primarily controlling discovery become defensible only through genuine usefulness, as access becomes programmable.
This reflects a universal pattern: when ecosystems gain automation layers, middlemen persist through value creation, not positioning. The channel doesn't disappear — it earns its place or loses it to the math.