Agentic Commerce, Retail Media, and Subscription Models

Executive Summary

The retail landscape is entering a period of fundamental transformation driven by three converging trends: agentic commerce, retail media networks, and subscription-based business models. For startup founders building technology solutions for fashion retailers, these emerging trends represent extraordinary market opportunities with substantial addressable markets and accelerating adoption curves.

McKinsey forecasts that agentic commerce—where AI agents autonomously shop, negotiate, and transact on behalf of consumers—could generate up to $1 trillion in orchestrated U.S. retail revenue by 2030, with global projections reaching $3 trillion to $5 trillion. This isn’t an incremental evolution of eCommerce; it represents a fundamental restructuring of how products are discovered, evaluated, and purchased. Adobe data shows traffic to U.S. retail sites from generative AI browsers and chat services increased 4,700% year-over-year in July 2025, whilst more than half of consumers anticipate using AI assistants for shopping by the end of 2025.

Simultaneously, retail media networks are creating new revenue streams as fashion retailers monetise their first-party data and customer attention, whilst subscription commerce—projected to reach $1.5 trillion in 2025 with 18% annual growth—offers predictable recurring revenue models. These trends aren’t isolated; they represent an interconnected transformation of retail business models, customer relationships, and revenue generation strategies.

This article examines the technical infrastructure requirements, strategic implications, and specific startup opportunities across these three emerging trends, with particular focus on fashion retail’s unique requirements and challenges.

Agentic Commerce: The Fundamental Restructuring of Shopping

Understanding the Paradigm Shift

The emergence of agentic commerce represents a paradigm shift as significant as the introduction of eCommerce itself. Unlike traditional eCommerce, where humans actively browse, compare, and purchase, agentic commerce enables AI agents to perform these tasks autonomously. ChatGPT now has more than 800 million weekly active users, and half of all consumers now use AI when searching the internet. What begins as AI-mediated discovery increasingly extends to execution, with agents comparing options, assembling baskets, and completing checkout via emerging payment protocols.

The adoption curve is accelerating rapidly. Adobe data shows traffic to U.S. retail sites from generative AI browsers and chat services increased 4,700% year-over-year in July 2025. More than half of consumers anticipate using AI assistants for shopping by the end of 2025, according to Adobe research. OpenAI’s Operator, launched in January 2025 and integrated into ChatGPT, enables users to automate tasks from product discovery through purchase completion. In late 2025, OpenAI announced an Agentic Commerce Protocol, co-developed with Stripe, allowing users to complete purchases within ChatGPT without leaving the conversation.

How Agentic Commerce Works

The system operates through two primary interaction models that fundamentally change retailer-customer relationships:

Consumer-to-Merchant (C2M): A consumer’s personal AI agent acts as their proxy, interacting with merchant agents to fulfil requests. For example, a shopper might instruct their agent: “I’m attending a wedding in Scotland in May and need an outfit suitable for unpredictable weather that fits my minimalist aesthetic.” The agent learns the user’s style, budget, and sizing, then evaluates options across merchants—autonomously interacting with inventory systems to confirm availability and payment agents to complete purchases.

Merchant-to-Merchant (M2M): Retailers’ AI agents interact with other merchant agents to extend capabilities beyond their own operations. If a consumer requests a product not in a retailer’s catalogue or currently out of stock, the merchant agent could interact with other retailers’ agents to source the item and complete the transaction—creating a frictionless shopping experience that prioritises customer needs over individual retailer boundaries.

The Infrastructure Requirements

Agentic commerce depends on new technical foundations that fashion retailers must implement urgently:

Structured Product Data: Rich, machine-readable product information becomes critical. AI agents cannot parse flashy landing pages—they require clean, structured data covering specifications, materials, sustainability credentials, sizing, availability, and pricing. For fashion specifically, this means semantic markup for fit characteristics, fabric composition, care requirements, and styling attributes.

Open APIs: API performance becomes as crucial as visual design. Agents need reliable, fast access to product catalogues, inventory systems, and checkout capabilities. The traditional emphasis on frontend user experience shifts to backend API reliability, response times, and data completeness.

Agent-Ready Authentication: New payment protocols enabling autonomous transactions. Mastercard’s Agent Pay technology and similar systems allow verified AI shopping agents to make transactions on behalf of consumers. Visa and Stripe have introduced agent toolkits supporting programmable, autonomous payments. These systems require identity verification, transaction authorisation frameworks, and fraud prevention adapted for non-human actors.

Conversational Interfaces: The traditional product detail page gives way to conversational experiences powered by large language models. Retailers need systems that can engage in natural language dialogue about products, answer complex questions about fit and styling, and adapt recommendations based on conversational context.

Strategic Implications for Fashion Retailers

BCG warns that without swift intervention, retailers risk being reduced to background utilities in agent-controlled marketplaces, losing direct customer relationships and brand engagement. The shift from human-centric browsing to algorithm-centric discovery means retailers must fundamentally rethink their digital strategies:

Optimise for AI Discovery: Just as retailers once optimised for search engines, they must now ensure their products are discoverable by AI agents. This requires semantic markup, structured data standards, and integration with agent platforms like ChatGPT, Perplexity, and Google Gemini. Fashion retailers need to think about how AI agents will interpret and compare their products against competitors.

Rethink Merchandising: Visual merchandising matters less than data richness. Product descriptions must be comprehensive enough for agents to understand nuances—fabric weight, fit characteristics, styling versatility—that humans would assess visually. The emphasis shifts from beautiful photography to comprehensive, structured product attributes.

Develop Merchant Agents: Fashion retailers need their own AI agents capable of engaging with consumer agents, negotiating on inventory allocation, suggesting complementary products, and personalising offers in real-time based on conversation context. This isn’t optional infrastructure; it’s core to competing in agent-mediated commerce.

Enable Dynamic Pricing: Agents can compare prices across retailers instantaneously and may even negotiate on behalf of consumers. Retailers need systems supporting dynamic pricing strategies that remain competitive whilst protecting margins. The transparency of agent-driven comparison shopping eliminates many traditional pricing strategies.

Startup Opportunities in Agentic Commerce Infrastructure

The agentic commerce infrastructure stack presents extraordinary opportunity for startups addressing fashion-specific requirements:

1. Agent-Ready Product Data Platforms

Solutions that transform existing product content into structured, machine-readable formats optimised for agent consumption. Fashion-specific semantic markup covering fit, material composition, care requirements, sustainability attributes, and styling characteristics. These platforms should integrate with existing PIMs (Product Information Management systems) whilst outputting agent-optimised data formats.

Market Opportunity: Every fashion retailer needs this infrastructure. Edgar Dunn & Company estimates the global total addressable market for agentic commerce transactions to reach approximately $136 billion by 2025. The product data infrastructure enabling this market represents a substantial opportunity.

2. Merchant Agent Development Frameworks

Pre-built conversational commerce agents specifically designed for fashion retail, with domain knowledge about sizing, styling, and product recommendations. These platforms should integrate with existing eCommerce systems whilst providing agent-native interfaces. Fashion-specific training on style terminology, seasonal trends, and outfit coordination creates defensible differentiation.

Market Opportunity: Fashion retailers lack the AI expertise to build merchant agents internally. Frameworks that enable rapid deployment with fashion-specific capabilities can capture significant market share before horizontal platforms adapt.

3. Agent Analytics and Optimisation

Tools helping fashion retailers understand how AI agents discover and evaluate their products, with recommendations for improving agent-mediated conversion rates. This represents a new category of commerce analytics focused on algorithm behaviour rather than human behaviour. Metrics include agent discovery rates, comparison patterns, selection criteria, and conversion drivers.

Market Opportunity: As agent-mediated shopping grows to 50%+ of eCommerce, retailers need visibility into this new customer journey. Analytics platforms providing this insight are essential infrastructure.

4. Payment and Checkout Orchestration

Middleware solutions enabling seamless integration with emerging agent payment protocols (Mastercard Agent Pay, Stripe agent toolkit, Visa’s agent standards) whilst maintaining fraud prevention and compliance requirements. These solutions must handle delegated authorisation, transaction limits, and verification workflows adapted for agent-initiated purchases.

Market Opportunity: Every transaction in the $136 billion 2025 TAM requires payment orchestration. Solutions providing security, compliance, and multi-protocol support capture payment processing margins.

5. Cross-Platform Agent Integration

Unified platforms managing product data syndication across multiple agent ecosystems—ChatGPT, Google Gemini, Perplexity, and emerging agent platforms—ensuring consistent representation and availability. As the agent ecosystem fragments across multiple platforms, retailers need unified management tools.

Market Opportunity: Fashion retailers selling across multiple channels already understand multi-channel management complexity. Agent platforms represent new channels requiring similar orchestration, creating addressable market amongst omnichannel retailers.

Market Sizing and Timing

McKinsey argues the transition to agentic commerce could unfold faster than previous digital shifts because agents operate on existing commerce infrastructure rather than requiring entirely new systems. For fashion retailers, the category’s visual and emotional components might seem incompatible with algorithm-driven shopping. However, the data suggests otherwise—fashion is amongst the earliest categories showing agent-mediated purchasing behaviour.

The strategic imperative is clear: fashion retailers and their technology partners must become “agent-ready” immediately. Those treating agentic commerce as a distant future scenario risk irrelevance as shopping behaviour fundamentally shifts from human-driven browsing to agent-mediated discovery and purchasing. For startups, the opportunity lies in building the infrastructure enabling fashion retailers to participate effectively in this new commerce paradigm.

Retail Media Networks: Monetising First-Party Data and Customer Attention

The Opportunity

The convergence of retail media networks, creator economy monetisation, and traditional eCommerce is creating new revenue streams for platforms that can orchestrate these elements effectively. Fashion retailers with significant traffic are building advertising businesses, monetising first-party data and customer attention.

Amazon’s advertising business demonstrates the model’s potential—generating billions in high-margin revenue by allowing brands to advertise on its platform using Amazon’s rich first-party data for targeting. Fashion retailers with established traffic and customer relationships can replicate this model, creating advertising inventory across their digital properties.

Infrastructure Requirements

Advertiser Self-Service Platforms: Fashion retailers need systems allowing brands (both internal product lines and external partners) to create, manage, and optimise advertising campaigns. These platforms must provide intuitive interfaces for campaign creation, budget management, and performance monitoring.

Campaign Management Tools: Backend systems managing ad inventory, bidding mechanisms, placement optimisation, and delivery across multiple touchpoints (homepage, category pages, search results, email, mobile app).

Measurement Solutions: Attribution and analytics infrastructure demonstrating advertising ROI. Fashion retailers must prove advertising effectiveness to command premium rates from brands.

Startup Opportunity

Retail media network infrastructure specifically designed for fashion retailers, including advertiser self-service platforms, campaign management tools, and measurement solutions. This represents significant opportunity as fashion retailers seek to replicate Amazon’s advertising success.

Differentiation: Fashion-specific advertising formats (lookbook placements, outfit completion suggestions, style-based targeting), integration with fashion calendars and seasonal trends, and creator partnership management for influencer advertising.

Market Opportunity: As fashion retailers recognise retail media as high-margin revenue streams, infrastructure providers enabling rapid deployment capture recurring revenue from advertising platform fees and transaction commissions.

Subscription and Membership Models: Predictable Revenue and Customer Lifetime Value

The Opportunity

Subscription commerce, projected to reach $1.5 trillion in 2025 with 18% annual growth, offers fashion retailers predictable revenue and improved customer lifetime value. However, implementation requires sophisticated subscription management infrastructure that most retailers lack internally.

Fashion subscription models include:

• Rental services: Continuous access to rotating wardrobes
• Styling services: Curated selections delivered regularly
• Replenishment subscriptions: Basics and essentials auto-delivered
• Membership programmes: Premium services and exclusive access
• Surprise boxes: Discovery-focused monthly deliveries

Infrastructure Requirements

Membership Programme Management: Systems handling subscription tiers, benefit provisioning, payment processing, and customer lifecycle management. Fashion retailers need flexibility to experiment with different subscription structures and pricing models.

Subscription Box Fulfilment: Logistics infrastructure supporting regular shipments, preference management, and returns processing for subscription-based inventory.

Customer Lifecycle Optimisation: Analytics and automation tools reducing churn, optimising pricing, managing pauses and restarts, and identifying upsell opportunities.

Startup Opportunity

Subscription commerce platforms for fashion, including membership programme management, subscription box fulfilment, and customer lifecycle optimisation tools. Fashion-specific functionality around styling services and wardrobe management provides differentiation.

Differentiation: Fashion-specific features including style profile management, outfit coordination across shipments, size and fit learning, seasonal preference adaptation, and sustainability tracking for rental models.

Market Opportunity: Fashion retailers recognise subscription revenue’s value but lack infrastructure for implementation. Platforms enabling rapid subscription launch with fashion-specific capabilities capture recurring revenue from subscription management fees.

Conclusion: Building Infrastructure for Retail’s Future

The convergence of agentic commerce, retail media networks, and subscription models represents the next wave of retail innovation. For startup founders, these trends create multiple entry points into fashion retail technology:

Immediate Opportunity - Agentic Commerce: With 4,700% traffic growth and 50%+ consumer adoption anticipated by year-end 2025, agentic commerce infrastructure needs are urgent. Fashion retailers lacking agent-ready systems risk displacement as consumer shopping behaviour shifts to AI agents. Five distinct infrastructure opportunities exist, each addressing critical gaps in the agent-mediated shopping stack.

Growing Opportunity - Retail Media: Fashion retailers are early in retail media adoption compared to categories like grocery and consumer electronics. Infrastructure providers enabling rapid retail media network deployment can capture significant market share whilst the category remains nascent.

Proven Model - Subscription Commerce: With $1.5 trillion in projected 2025 revenue and 18% growth, subscription commerce demonstrates validated consumer demand. Fashion-specific subscription infrastructure addresses known pain points in an established market.

Strategic Positioning: The most successful startups will build composable solutions that integrate across these trends. Agent-ready product data platforms that also syndicate to retail media systems, or subscription platforms with agent-mediated management capabilities, create stronger competitive positions than point solutions.

Urgency: McKinsey, BCG, and Google Cloud all emphasise the same message—the transition to agentic commerce will happen faster than previous digital transformations because it operates on existing infrastructure. Fashion retailers and their technology partners must act immediately. Those building the infrastructure enabling this transition will capture disproportionate value in the emerging retail landscape.

For founders targeting fashion retail, the opportunity to build venture-scale businesses addressing these emerging trends is compelling. The market timing is exceptional—early enough to establish category leadership, but late enough that market validation and customer urgency are established. The retailers that successfully navigate these transformations, supported by innovative technology partners, will define the future of fashion commerce.

Image courtesy of Unsplash

---

References

1. McKinsey & Company. (2025). The Agentic Commerce Opportunity: How AI Agents Are Ushering in a New Era for Consumers and Merchants.

2. Boston Consulting Group. (2025). Agentic Commerce is Redefining Retail - How to Respond.

3. Adobe. (2025). Consumer AI Shopping Behaviour and Traffic Analysis.

4. Renner, M. (2025). How Agentic AI Will Change Commerce As We Know It. Fortune.

5. Google Cloud. (2025). Agentic Commerce is Here: How Retailers Can Prepare for the New Shopping Era.

6. Mastercard. (2025). What is Agentic Commerce? Your Guide to AI-Assisted Retail.

7. Edgar Dunn & Company. (2025). Agentic Commerce: The Future of Payments.

8. Digital Commerce 360. (2025). McKinsey Forecasts Up to $5 Trillion in Agentic Commerce Sales by 2030.

9. Morgan Lewis. (2025). Agentic AI: An Agent of Change in the Ecommerce Space. Tech & Sourcing @ Morgan Lewis.

10. The Fashion Law. (2025). AI Goes Shopping: Inside the New Era of Agentic Commerce.

11. CIO. (2025). Agentic Commerce: How AI is Reinventing the Way Customers Discover and Buy.

12. OpenAI. (2025). Agentic Commerce Protocol. Co-developed with Stripe.

13. Stripe. (2025). Agent Toolkit for Agentic Commerce. Payment infrastructure for AI agent transactions.

14. Visa. (2025). Agent Payment Standards and Purchase Signal Protocol.

15. UBS. (2024). Subscription Economy Forecast Report.