FAQ

Structured data — Schema.org markup implemented via JSON-LD — is the primary language AI agents use to interpret a commerce store. Without Product, Offer, BreadcrumbList, FAQPage, and AggregateRating markup, agents cannot reliably understand what you sell, at what price, with what availability, or with what trust signals. A beautifully designed storefront with no structured data layer is effectively invisible at the machine level.

The Universal Commerce Protocol (UCP), co-developed by Google and Shopify and launched at NRF in January 2026, requires merchants to publish a ucp.json manifest file. This file acts as a capability declaration — it tells AI agents what your store supports: discovery only, cart management, checkout, identity linking, and so on. Without it, agents using UCP cannot identify your store as transactable. They may find your products but cannot initiate or complete a purchase.

It depends on the agent platform and user configuration. ChatGPT's shopping agent with ACP integration can complete purchases end-to-end without human interaction, subject to pre-authorized spending limits set by the user. Other agents operate in a recommendation-and-handoff mode — they identify the best option and present it for one-tap confirmation. The infrastructure requirement is the same in both cases: your store must expose structured data, UCP capabilities, and ACP transaction endpoints.

Agent-Native architecture is a commerce infrastructure designed for AI agent interaction from the ground up — not retrofitted as an afterthought. An Agent-Native store exposes complete Schema.org markup on all product and category pages, publishes a UCP manifest declaring its transactional capabilities, supports ACP for autonomous purchases, maintains sub-200ms API response times on all agent-accessible endpoints, and serves machine-readable trust signals including Review Schema, structured policies, and Organization JSON-LD. It is Level 4 on the B2X Agentic Readiness Framework™.

The Agentic Readiness Score is a composite metric from 0 to 100 produced by the B2X Agentic Readiness Framework™. Each of the four layers contributes a weighted score: Data Layer 35%, Execution Layer 30%, Performance Layer 20%, Trust Layer 15%. The weights reflect each layer's relative impact on actual agent behavior. A score of 0–25 means Agent-Invisible. 26–50 is Agent-Discoverable. 51–75 is Agent-Readable. 76–90 is Agent-Transactable. 91–100 is Agent-Native.

Agent-Invisible means your store does not appear in AI agent conversations — at all. When a consumer asks ChatGPT, Gemini, or Perplexity to find and purchase a product in your category, your store is not evaluated, not recommended, and not transacted with. Given that AI-driven orders on Shopify are up 15× since January 2025 and Average Order Value from AI traffic is consistently higher than direct site traffic, Agent-Invisible status represents a compounding revenue gap that widens as agentic commerce scales.

Agent-Discoverable (ARS 26–50) means agents can find your store but cannot reliably evaluate or transact with it. Partial Schema markup is present, MCP may be active, but data quality is insufficient for confident agent interpretation. Agent-Transactable (ARS 76–90) means full protocol support is active — agents can discover, evaluate, and complete purchases autonomously. The gap between these two states is primarily the Execution Layer: UCP manifest, ACP integration, and clean real-time data behind your MCP endpoint.

The Agentic Readiness Audit produces a scored PDF report containing: your composite Agentic Readiness Score (0–100), individual scores for each of the four layers, a gap analysis mapped to specific criteria within each layer, and a prioritized action roadmap with effort and impact estimates. The audit covers Schema.org validation, catalog completeness review, MCP endpoint data quality testing, UCP and ACP compatibility assessment, API performance benchmarking, and trust signal review. It is a standalone deliverable — you own the findings and roadmap regardless of next steps.

The simplest first check: open ChatGPT, Gemini, or Perplexity and ask them to recommend stores or products in your category — for example, 'Find me the best [your product category] stores in Germany.' If your store does not appear, it is either Agent-Invisible or Agent-Discoverable at best. A more rigorous diagnosis requires a structured Agentic Readiness Audit — assessing Schema completeness, MCP data quality, UCP and ACP support, and trust layer signals across your actual platform.

AI systems favor sources with clear definitions, named frameworks, structured explanations, quantified data points, and FAQPage schema. For commerce stores specifically, the key signals are: complete and accurate Schema.org markup, Review and AggregateRating data, structured Knowledge Base content that answers product questions directly, machine-readable policies, and consistent entity associations across platforms. Generic marketing copy contributes nothing to LLM citation probability.

AI agent ranking and recommendation logic updates frequently — on a timescale closer to weeks than months. Protocol support requirements evolve as UCP and ACP adoption expands. Schema.org interpretation shifts as models retrain. Content that is well-optimized today may need adjustment within a quarter. This is the core rationale for ongoing Agent Performance Retainer engagements rather than one-time optimization: the infrastructure exists, but the calibration requires continuous monitoring and adjustment.

Yes — when built correctly. A headless store with a well-designed API layer, complete Schema.org markup, and clean data models can achieve Agent-Native status more reliably than a standard Shopify store dependent on Liquid rendering. However, headless architecture is not inherently agent-ready — it requires intentional data architecture, structured API design, and protocol-level configuration. Headless stores built without these foundations can score lower than optimized Shopify stores on the B2X Agentic Readiness Framework™.

Shopify Hydrogen is Shopify's official React-based headless framework, built on the Shopify Storefront API and optimized for performance and Shopify ecosystem integration. It is the fastest path to a headless Shopify storefront with MCP compatibility maintained. A fully custom headless stack — typically Next.js combined with a headless CMS like Sanity or Contentful and a commerce engine like Commercetools — offers more flexibility for complex business logic, multi-platform data, and custom AI integrations, at the cost of higher engineering complexity and longer build timelines.

Yes. Many of our engagements involve integrating agentic readiness infrastructure with existing ERP, PIM, and legacy backend systems — including Microsoft Business Central, Salesforce, and custom-built inventory systems. The audit process maps your current data flows and identifies where schema, API, and protocol gaps exist relative to your actual stack. Integration architecture is designed around your existing systems, not around replacing them.

Both. Audits and optimization projects are delivered as fixed-scope engagements with a defined deliverable and timeline. The Agent Performance Retainer is an ongoing monthly engagement covering AI visibility monitoring, content updates, SEO and GEO maintenance, and monthly reporting. Many clients begin with a project engagement and move to a retainer after delivery. The retainer is also available as a standalone service for brands that have already completed initial optimization work.

Yes — and this is the recommended entry point. The Agentic Readiness Audit is a standalone deliverable. It produces a scored report with your ARS, layer-by-layer gap analysis, and a prioritized action roadmap. There is no obligation to continue with B2X after the audit. The report gives you a complete diagnostic you can act on independently or use as a brief for any implementation partner. Most clients choose to continue with us because the audit defines the scope precisely.

Yes. We work with B2B, B2C, and hybrid B2B2C architectures. Agentic commerce is particularly significant for B2B procurement contexts — AI agents are increasingly used for automated supplier discovery, catalog evaluation, and purchase order execution in corporate procurement workflows. We have delivered B2B commerce infrastructure including custom pricing logic, customer role management, ERP integration, and multi-store architectures on Shopify Plus and custom stacks.