Imagine a shopper who never browses aimlessly, never abandons their cart out of frustration, and always arrives with clear intent and a payment method already loaded. Now imagine that shopper is an AI agent acting on behalf of millions of consumers simultaneously.

That future is not a forecast. It is happening now — and it is happening faster than most enterprise leaders have adjusted to. Research from Boston Consulting Group and Commercetools, published within months of each other, arrives at the same conclusion from different angles: the retail industry is undergoing a structural transformation that redefines not just the shopping experience, but who — or what — constitutes a customer.

4,700%
YoY growth in GenAI-driven US retail traffic, July 2025 (Adobe)
73%
Of consumers already using AI in their shopping journey (Global Study, 2025)
25%
Projected share of online spending through AI agents by 2030 (Morgan Stanley)
96%
Of global retailers exploring or implementing AI agents (monday.com survey)

What Agentic Commerce Actually Means

The term gets used loosely. It is worth being precise. An AI agent in this context is not a chatbot that answers product questions. It is a system that acts — autonomously or semi-autonomously — to complete a goal. It researches options, filters by preference, compares prices, selects the best match, and completes a purchase — potentially without a human clicking a single button.

BCG frames the shift clearly: instead of consumers manually searching, comparing, and deciding, agents scan platforms, filter results against individual preferences, and make context-aware recommendations. As trust in these systems grows, they increasingly take on transactional tasks — checking delivery timelines, applying stored payment credentials, and completing purchases on behalf of users.

A retailer's most valuable customer might not be a human. Without intervention, retailers risk being reduced to background utilities in agent-controlled marketplaces.

The scale of early adoption data makes this more than theoretical. AI-referred traffic to US retail sites grew 805% year-over-year on Black Friday 2025. Shoppers arriving through AI channels spend 32% more time on site, browse 10% more pages, and bounce 27% less than traditional visitors. They arrive further down the decision funnel, already primed to buy. The problem for retailers is that the journey leading up to that moment — the discovery, the comparison, the consideration — happened somewhere else entirely.

The Seven Forces Shaping the Landscape

Commercetools identifies seven trends defining agentic commerce in 2026. Together they describe not a single disruption but an ecosystem in rapid formation.

01
GenAI Platforms Become Retail Channels
ChatGPT, Gemini, and Perplexity have moved from assistants to storefronts. Retailers that optimise only for traditional search are increasingly invisible where buying decisions are now being formed.
02
Discovery and Conversion Lead; Loyalty Follows
The earliest agentic capabilities cluster at the top of the funnel. Full end-to-end agent journeys are coming but the ecosystem is still maturing. The window to build presence in the discovery layer is now.
03
Purpose-Built Agents Outpace Monolithic Solutions
62% of organisations are experimenting with AI agents but only 23% are scaling. The pragmatic on-ramp is specialised agents — a reorder tool, a bundle builder, a shopping assistant — rather than a master agent orchestrating entire journeys.
04
Brand-Owned Conversational Commerce Gets an Upgrade
Brand-owned AI agents can guide discovery, curate offers, manage carts, and navigate loyalty programs — creating depth of interaction no traditional chatbot could provide.
05
AI Reshapes Back-Office Operations
The transformation is not only consumer-facing. Internal productivity tools, AI-assisted forecasting, and autonomous workflow orchestration are widening the gap between early adopters and laggards.
06
B2B Commerce Is the Next Frontier
Gartner projects 90% of B2B buying will be AI agent-intermediated by 2028. The structural complexity of B2B makes it the sector with the greatest near-term ROI from agentic deployment.
07
Trust Becomes the Ultimate Differentiator
As agents gain authority to act on behalf of consumers, the brands that win will be those who embed transparency, consent controls, and audit trails at the foundation of their agentic experiences.

The Risks Retailers Cannot Ignore

BCG's analysis is direct about the structural threat. As GenAI platforms become the default entry point for online shopping, retailers face disintermediation unlike anything the e-commerce era has produced. Not just traffic loss — loss of the entire decision-making context.

In traditional e-commerce, retailers see everything: impressions, clicks, dwell time, add-to-cart events, funnel drop-offs. In agent-mediated commerce, that behavioural data stream starts at the add-to-cart moment. The discovery, browsing, and consideration phase now lives inside ChatGPT or Gemini, invisible to the retailer. First-party data that powers personalisation, loyalty programs, and retail media revenue evaporates.

AI agents also behave differently from human shoppers. They prioritise price, ratings, delivery speed, and inventory availability over brand familiarity. Loyalty built through years of brand investment becomes less durable when the purchasing decision is made by a system optimising for utility.

How to Respond: Three Pillars

BCG's framework for navigating this shift organises around three investment priorities. Together they describe what it means to be genuinely AI-ready as a retail organisation.

Pillar One
Win with Third-Party Agents
Traditional SEO gives way to Generative Experience Optimisation (GXO) — structuring product content, metadata, and data assets to be authoritative, semantically rich, and machine-readable. AI search ad spending in the US alone is projected to reach $26 billion by 2029.
Pillar Two
Build Retailer-Owned Agentic Experiences
Three types of proprietary agents give retailers a differentiated position: brand agents with deep domain expertise; workforce agents that transform employee productivity; and partner agents that automate supplier and merchant relationships.
Pillar Three
Build Agentic Foundations
Adopt the Model Context Protocol (MCP) as a standard for AI-to-system integration. Build measurement frameworks to track brand visibility across GenAI platforms. Establish AI governance with clear roles and risk management. Invest in talent designed to operate at AI speed and scale.

What This Means for the C-Suite

The implication of all this research converges on a single strategic question: Is your organisation positioning itself to be found, chosen, and transacted with by AI agents — not just human shoppers?

That question touches every layer of the enterprise. Product data architecture determines whether agents can surface your inventory. Pricing strategy determines whether agents recommend you or a competitor. Governance frameworks determine whether your brand behaves consistently when an AI is acting in its name.

The organisations that will lead in this era are not necessarily those with the largest AI budgets. They are the ones that move from experimentation to deliberate architecture — making clear decisions about which agents to build, which platforms to prioritise, and which foundations to put in place now, while the standards and protocols are still being written.

The window to lead is rapidly closing, but for those bold enough to act, the upside is transformative. Those who wait risk becoming invisible — disintermediated from the customers they once knew.