The rise of agentic commerce: how AI will reshape merchant strategy

With agentic commerce on everyone’s lips, Sameer Verma from Payments Genes examines it from a merchant perspective, reviewing user journeys, challenges, operating models, and value creation opportunities.

Introduction

Commerce has reached a full circle moment. Prior to the internet, buyers and sellers negotiated directly. Today, purchasing decisions are shaped by interface design, brand perception, and checkout friction. Ecommerce shifted interactions from human-to-human to human-to-machine. AI agents represent the next inflection point, acting on behalf of consumers to search, compare, and transact across thousands of merchants in seconds. While this promises speed and convenience for shoppers, it introduces significant risks for merchants, including loss of brand loyalty, reduced customer proximity, and increased commoditisation.
This article examines agentic commerce from a merchant perspective, reviewing user journeys, challenges, operating models, and value creation opportunities.

1) Agentic commerce user journey

Whether agentic commerce becomes a powerful acquisition channel or cannibalises merchant sales depends on technical foundations such as APIs, data pipelines, and transaction endpoints. Regardless, customer journeys will be significantly compressed compared to traditional ecommerce.

Key differences between human and machine buying behaviours include:

1. Information absorption and processing: AI agents can rapidly consume and compare vast product datasets, enabling detailed assessments of price, quality, warranties, and features.
2. Speed and scale: Agents operate continuously and in parallel across hundreds of funnels. For example, when booking flights, they can simulate full checkout flows across airlines to surface true total costs, including add-ons.
3. Insensitivity to intangibles: Unlike humans, agents are less influenced by brand emotion, imagery, or website experience, focusing instead on measurable indicators of product quality and value.

 

2) The promise and the challenges

McKinsey estimates that agentic commerce could generate up to USD 1 trillion in US B2C retail revenue by 2030, with global projections of USD 3–5 trillion. Its impact is expected to mirror earlier web and mobile commerce shifts. However, current infrastructure presents significant challenges:

1. Agent identification: Existing fraud systems struggle to distinguish legitimate AI agents from malicious bots, risking blocked transactions.
2. Loss of consumer proximity and data: As agents mediate purchases, they control customer interactions and insights, weakening merchant differentiation and loyalty.
3. Fragmented payments: Lack of payment standardisation, especially in Europe, complicates agent-based transactions.
4. Marketplace disintermediation: AI agents may bypass traditional marketplaces by independently discovering and comparing sellers.
5. Long-tail exclusion: Merchants with basic HTML storefronts and no API connectors or feeds risk being locked out of agentic ecosystems.
6. Compliance and risk: New fraud patterns and exposure to raw payment data may increase PCI and regulatory risk.

 

3) Operating models and use cases

Agentic commerce is still nascent, with multiple models emerging. Most current implementations support live product search and real-time payment authentication through assisted checkout. Over time, fully autonomous transactions will become possible through digitally signed user authorisations defining budgets, timing, and payment methods.

Embedded payments in AI chatbots are already enabling seamless transactions. Today’s primary use case is product and deal discovery, where agents aggregate prices, pre-fill carts, and await manual approval. Broader use cases include concierge shopping, complex travel planning, and targeted purchasing based on predefined criteria.
In enterprise settings, procurement agents can enforce spending policies using virtual cards, automatically trigger purchase orders, and process invoices once inventory thresholds are met.

4) Enabling agentic commerce

Technology providers across payments, AI, and infrastructure are developing protocols to support agentic commerce. While the growing number of standards may create confusion, they address different parts of the value chain. For example, some protocols enable programmable payment tokens, while others focus on verifying user intent or distinguishing trusted agents from bots through cryptographic signatures.

Common security approaches include:

• Verification of user intent through cryptographic signatures or tokens;
• Multi-layer authentication via as biometrics, passkeys, and 2FA;
• Non-repudiable audit trails for compliance and dispute resolution.

Agentic commerce enablement:

Agentic AI Implementation relies on support infrastructure. To enable agentic commerce, merchants are likely to adopt one of three implementation models:

1. Light adapters: Connecting agent traffic through existing gateways with minimal operational change.
2. Hybrid models: Separating agent and human traffic while protecting existing conversion flows.
3. Full offloading: Creating agent-specific catalogues and landing environments optimised for machine interaction.

 

5) Creating differential value in agentic commerce

For most merchants, enabling agentic commerce will be a defensive strategy. However, early leaders can gain advantage by using it to create differentiated value. Strategies include:

1. Personalised offers: Using richer demand data to deliver tailored bundles and incentives that improve conversion and loyalty.
2. Smart inventory and pricing: Leveraging SKU-level insights to optimise stock and pricing based on real-time demand signals.
3. Concierge services: Integrating AI-driven assistance within merchant sites to guide customers before and after purchase.
4. Agentic landing pads: Designing optimised environments for agent interaction and seamless handoffs to human users with preserved context.

Agentic commerce is still in its early stages, and its ecosystem is rapidly evolving. Ongoing observation and experimentation will be critical as merchants and payment providers adapt to this new paradigm.

About author

 

Sameer Verma, Manager at PaymentGenes Consulting, advises global enterprises and fintechs on payment optimisation, go-to-market, and product strategies. With rich experience in the payments-focused M&A and strategy advisory, he has worked across the transaction and banking infrastructure. His work increasingly sits at the intersection of payments and AI, translating technological shifts such as agentic commerce into practical merchant and platform strategies.

 

 

About Payment Genes

We combine deep sector expertise with best-in-class management consulting to help our clients create value with payment innovations. Whether you are an enterprise merchant, PSP, acquirer, issuer, payment scheme, incumbent bank, or fintech, we leverage our global expertise from across the value chain to research, benchmark, strategise, design, perform the vendor selection, and implementation for your next-generation payments solutions. 

Our Consultancy has supported leading brands with their payments projects across the globe.