Something fundamental is changing in how consumers shop online. I've spent over 20 years watching digital commerce evolve, from the early days of SEO to mobile-first online shopping to the current AI revolution.
But what's happening now isn't a gradual evolution. Agentic commerce represents a fundamental shift in the relationship between buyers, brands, and the technology that connects them.
Picture this: instead of typing keywords into search engines, scrolling through endless product pages, and comparing reviews across multiple tabs, your customer simply tells AI agents what they need. These AI agents ask clarifying questions about budget, preferences, and use cases.
They research options across the web, synthesize reviews, compare specifications, and present curated recommendations. Then they complete purchases without the customer ever leaving the conversation.
This isn't a demo at a tech conference.
AI agents acting on behalf of consumers are doing this right now in ChatGPT, Perplexity, and Google Gemini. The commerce ecosystem is transforming before our eyes, and agentic commerce is leading that transformation.
The Market Opportunity Behind Agentic Commerce
The numbers tell a compelling story about this new era of digital commerce. McKinsey projects agentic commerce could generate up to $1 trillion in orchestrated U.S. retail revenue by 2030, with global projections reaching $3 trillion to $5 trillion. McKinsey describes agentic commerce as a "seismic shift" that moves us toward a world where AI agents anticipate consumer needs, navigate shopping options, negotiate deals, and execute transactions on behalf of users.
Morgan Stanley predicts nearly half of all U.S. e-commerce shoppers will use AI agents by 2030, potentially adding $115 billion in total online shopping spending. Their analysis suggests agentic commerce gross merchandise value could reach between $190 billion and $385 billion by the end of the decade.
The adoption curve is accelerating rapidly. Adobe reports AI-driven traffic to U.S. retail sites surged 4,700% year-over-year in July 2025. These aren't just visits. AI-referred shoppers are 32% more engaged, view 10% more pages, and have a 27% lower bounce rate than those from traditional channels. Commerce agents are changing consumer behavior in real time.
For marketing leaders at consumer brands, this shift in agentic commerce demands attention. The brands that understand this transition early will capture disproportionate advantage. Those who wait risk becoming invisible in a new era where traditional SEO and paid search matter less than they did yesterday.
What Does Agentic Commerce Mean?
Let me be precise about what agentic commerce refers to, because "AI in e-commerce" has become a catch-all phrase that obscures more than it reveals.
Agentic commerce is distinct from the chatbots you've seen on retail websites or the product recommendation engines that power "customers also bought" suggestions. Those applications use traditional AI to enhance existing shopping experiences. Agentic commerce replaces the experience entirely with AI-powered agents that work autonomously.
The defining characteristic is agency. AI agents don't just respond to queries or surface options. Intelligent agents act on the user's behalf through multi-step tasks of reasoning and action. These AI-powered agents research, compare, evaluate trade-offs, and increasingly complete purchases with minimal human input. They enable agents to dynamically adapt to user intent and consumer intent throughout the shopping journey.
Think of the difference this way: a traditional AI chatbot answers the question "What are good running shoes for beginners?" Agentic AI systems built on advanced AI and large language models ask follow-up questions about your running goals, typical distance, foot type, and budget. These AI agents then search across retailers, read expert reviews, synthesize user feedback, identify the best options for your specific situation, and can complete the purchase within a single conversation, executing purchases on the user's behalf.
This fundamental shift from reactive assistance to proactive assistance is why agentic commerce matters for brands. When AI agents acting on behalf of users become the primary interface between consumers and digital commerce, the rules of product discovery, consideration, and conversion fundamentally change.
The consumer isn't visiting your website, browsing your category pages, or clicking your paid search ads. They're having a conversation with consumer agents or shopping agents that may or may not recommend your products based on criteria you don't fully control and through AI channels you may not be optimizing for.
How AI Shopping Agents Actually Work
Understanding the mechanics of agentic commerce helps clarify why this shift matters strategically, even if you never touch the technical implementation yourself. AI agents rely on specific infrastructure to work effectively.
Product Data Is the New Currency
The cornerstone of AI-powered product discovery in agentic commerce is structured data. This includes comprehensive product feeds that provide AI agents with everything they need to know about your products in machine-readable format.
This isn't entirely new. You've probably maintained Google Shopping feeds for years. But AI platforms require significantly more detail than traditional product feeds. OpenAI's Product Feed Specification exemplifies this expanded requirement for enabling agents to discover and recommend products.
Beyond the basics like title, description, price, and availability, AI feeds include performance signals like popularity scores and return rates, custom variant dimensions beyond standard color and size, compliance metadata for regulatory requirements, and rich media including videos and 3D models. Product data must be comprehensive for commerce agents to make good recommendations.
Here's what makes this strategically important for agentic commerce: unlike traditional search where your website content, backlinks, and domain authority influence rankings, AI shopping agents treat the product feed as the primary authority on your brand and products. The feed isn't just another signal. It's the foundation of your visibility in this new commerce ecosystem where AI agents make decisions.
Character limits are strict. Product titles max out at 150 characters, product descriptions at 5,000. Updates can occur as frequently as every 15 minutes for real-time insights on pricing and inventory accuracy. Your product data strategy becomes a critical marketing asset in the age of agentic commerce.
Semantic Search Enables Agents to Understand Consumer Intent
Traditional search engines match keywords. AI shopping powered by generative AI and agentic AI understands consumer intent and user intent at a deeper level.
When a consumer tells ChatGPT they need "a durable carry-on under $300 that fits in overhead bins on regional jets," intelligent agents don't just search for those exact words. AI agents achieve a deep understanding of functional requirements: size constraints for smaller aircraft, durability expectations, price ceiling, and the implied use case of frequent travel.
This semantic understanding is powered by embedding models using natural language processing that convert product descriptions and user queries into mathematical representations. Products that functionally match what the consumer needs surface even if they don't contain the exact keywords in the query. AI agents acting on behalf of users can parse natural language requests and find the right products.
The business implication for agentic commerce: keyword stuffing and traditional SEO tactics become less effective. What matters is whether your product data comprehensively and accurately describes what your product actually does, who it's for, and how it solves problems. Clarity and completeness trump keyword optimization when AI agents are making consumer purchasing decisions.
Why Real-Time Accuracy Matters More Than Ever
AI shopping agents make specific promises to consumers: this product is available, at this price, with these specifications. When that information is wrong, trust erodes in both your brand and the AI platforms where commerce agents operate.
This creates strong incentives for AI platforms to prioritize merchants with accurate, frequently updated data. If your inventory feed shows products as available that are actually out of stock, or prices that don't match your website, you're creating friction that AI agents are designed to eliminate.
The practical takeaway for agentic commerce: data hygiene and real-time synchronization aren't just operational concerns. They're competitive advantages for operational efficiency in AI-driven discovery, where agents need secure transactions and accurate information.
The Major Agentic Platforms and What They're Building
The AI shopping landscape is consolidating around a few major agentic platforms, each with distinct approaches to how consumers interact with commerce agents. Understanding these differences helps prioritize where to focus optimization efforts for agentic commerce.
ChatGPT and the Agentic Commerce Protocol
OpenAI has moved aggressively into commerce. Their shopping research feature, powered by specialized AI agents trained specifically for shopping tasks, now supports Instant Checkout through the Agentic Commerce Protocol developed with Stripe. U.S. users can complete purchases directly within the chat interface, with Etsy sellers already integrated and Shopify merchants, including brands like Glossier, SKIMS, and Vuori, rolling out.
The agent payments protocol and transaction flow maintain merchant control. You remain the merchant of record with complete control over orders, payments, fulfillment, and customer relationships. ChatGPT acts like a personal shopper, handling product discovery and checkout, while you handle everything post-purchase using your own payment credentials.
Current capabilities support only single-item purchases from U.S. merchants. Multi-item carts and international transactions aren't yet supported. This is early infrastructure for enabling agents to execute agent-initiated transactions, not a mature capability.
What matters for marketers in agentic commerce: ChatGPT's product recommendations are organic and unsponsored, ranked purely on relevance. There's no paid placement yet. This means feed optimization and product data quality directly impact visibility without the ability to buy your way to prominence in AI agents' recommendations.
Perplexity as Real-Time Research Engine
Perplexity has positioned itself as the research-first shopping platform within the agentic commerce ecosystem. Unlike ChatGPT's conversational approach, Perplexity emphasizes real-time web search combined with product discovery. Their "Buy with Pro" feature enables Pro subscribers to complete one-click checkout directly within the platform, allowing AI agents to complete purchases seamlessly.
The platform offers a "Snap to Shop" feature that lets users upload photos of products to initiate searches. This visual search capability, similar to Google Lens, represents another entry point for consumers alike to discover products through AI agents.
Their merchant program is currently free to join and doesn't take commissions from affiliates. Merchants who provide detailed product data are more likely to appear in recommendations from their commerce agents. This early-mover opportunity in agentic commerce won't last forever.
Google Cloud and Agentic Checkout Features
Google has launched agentic checkout within Google Search, including in AI Mode. The feature works with eligible merchants like Wayfair, Chewy, Quince, and select Shopify merchants through their agentic commerce infrastructure.
Google Cloud has published extensive guidance for retailers preparing for agentic commerce. Their approach leverages its Shopping Graph, which includes more than 50 billion product listings with 2 billion updated every hour for AI agents to access. Users can track an item's price and opt to have Google's AI agents purchase it automatically when the price falls within their budget using Google Pay.
Google Cloud's Conversational Commerce agent, built on Gemini models, achieves a precise understanding of consumer intent, enabling agents to guide shoppers from initial discovery to completed purchase. This represents a significant investment in the agentic commerce future.
The Future Trajectory of AI Agents in Agentic Commerce
Understanding the likely evolution of agentic commerce helps prioritize current investments and stay ahead of coming changes in how AI agents transform shopping.
AI Agent Capabilities Are Improving Exponentially
Research from METR shows that the duration of tasks that large language models can reliably complete has approximately doubled every 7 months since 2019. In 2019, leading models could handle tasks that required only a few seconds of human effort. By early 2025, advanced AI models like Claude can complete tasks that would take a skilled human around 50 minutes.
If this trend continues, AI agents may handle multi-hour shopping research and comparison tasks autonomously in the near future. This includes complex purchases involving extensive research, multiple options evaluation, and nuanced trade-off analysis. Agentic commerce will enable agents to execute increasingly sophisticated multi-step tasks without human intervention.
Autonomous Shopping Agents Are Coming
The evolution toward fully autonomous shopping agents represents the logical endpoint of current agentic commerce trends. For routine purchases such as household staples, repeat orders, and commodity items, consumers may increasingly delegate entirely to their own agents, operating with preset preferences and spending limits.
For high-consideration purchases, human intervention remains likely, but AI agents handle all research and option narrowing. Human shoppers focus on final decisions, while commerce and shopping agents handle the heavy lifting of product discovery and comparison.
The strategic implication for agentic commerce: brands need to think about appealing not just to human shoppers but to the AI agents working on behalf of users.
What signals indicate quality and fit-for-purpose to intelligent agents evaluating thousands of options? Competitive pricing matters for agentic commerce, but so does brand loyalty and precise product data that AI agents can parse.
Regulation of Agentic Commerce Is Coming
California's Transparency in Frontier Artificial Intelligence Act, signed in September 2025, signals increasing regulatory attention to AI agents and agentic commerce. The law establishes the first comprehensive state framework for transparency, safety, and accountability in AI development.
Key concerns for agentic commerce include advertising disclosure requirements for AI agents' recommendations, consumer protection regulations for automated commerce where agents complete purchases, data privacy compliance across jurisdictions, and ethical considerations for AI-generated product suggestions. Responsible AI practices will be essential as third-party agents gain more autonomy in executing purchases.
Brands operating in agentic commerce should monitor regulatory developments and build compliance considerations into their strategies early. The model context protocol and other agent standards will likely face scrutiny as agentic commerce scales.
The Bottom Line on Agentic Commerce
Agentic commerce represents a fundamental reimagining of how consumers interact with brands to find, evaluate, and purchase products through AI agents. The shift from traditional online shopping to conversational discovery, where AI agents act on behalf of consumers, isn't merely a technological upgrade. It's a structural change in the relationship between brands and buyers in the digital commerce landscape.
The market opportunity for agentic commerce is substantial: trillions of dollars in global economic value as AI agents become the new gatekeepers of commerce. Shopping agents, commerce agents, and consumer agents are already reshaping how consumers interact with brands and how consumer experience unfolds.
But the window for establishing competitive advantage in agentic commerce is relatively narrow. The brands that understand this transition early will capture disproportionate value.
The imperative for marketing leaders is clear: begin experimenting with agentic commerce now while the market is still forming, but maintain realistic expectations about current limitations and near-term capabilities. The winners in this agentic commerce transition will combine superior product data quality with strategic positioning, user-centric thinking, and adaptive approaches that evolve as AI platforms and AI agents mature.
The shift to agentic commerce is happening. Intelligent agents are already changing consumer behavior. The question isn't whether to engage with agentic commerce, but how quickly and how strategically you can position your brand to win in a world where AI agents complete purchases on behalf of users.