Skip to content
Universal Commerce Protocol Protocol registry / en

Methodology

UCP readiness audit
in four phases.

Mapping, scoring, arbitration, roadmap. Each phase delivers a standalone, reusable artifact that can be replicated in 12 months.

Duration

4 to 6 weeks

Format

Hybrid: async + 2 workshops

Deliverable

Written report + exec summary

Pricing

Custom, scoped

Phase I

Mapping

Indicative duration: 5 to 10 days

Inventory of catalog objects, endpoints, payment rails and existing policies. No recommendations at this stage: we describe what exists.

Inputs

  • PIM / feed access
  • Policy inventory
  • E-commerce stack documentation
  • SEO reports (12 months)

Output

Map of the four UCP layers in their current state, with documented gaps and gray areas.

Phase II

Scoring

Indicative duration: 3 to 5 days

Each layer receives a score based on a 40-point grid. Scoring is replicable: the same grid is applied before and after.

Inputs

  • Phase I results
  • UCP 40-point grid
  • 30-min interviews with 5 teams

Output

UCP score per layer (0-100), internal cross-team benchmark, three identified asymmetric risks.

Phase III

Arbitration

Indicative duration: 1 workshop (3 hours)

Structured workshop with leadership: we arbitrate, we do not present. Each decision is documented with its cost, expected gain and uncertainty level.

Inputs

  • UCP score
  • Data + tech budget constraints for 2026
  • Commercial priorities 12-18 months

Output

Arbitration matrix validated by the executive team, with high / medium / hold scenarios.

Phase IV

Roadmap

Indicative duration: 2 to 3 days

Operational action plan with 3 horizons, per-layer indicators, and a recalibration checkpoint every 90 days based on market signals.

Inputs

  • Arbitration matrix
  • Available internal resources
  • Deployment windows

Output

6 / 12 / 18-month roadmap, steering indicators and governance rituals.

What the mapping phase actually looks for

The grid is not arbitrary. It follows what the UCP specification asks a merchant to publish. A readiness audit is the exercise of holding your current setup against those requirements, one by one.

The profile and capability negotiation

UCP works by negotiation: a business publishes a profile describing its capabilities and services, and the platform discovers dynamically what it can do with that business. So the audit asks less "do you have an API" than "what are you declaring you can do, and is it true". Capabilities are named using reverse-domain notation, which lets any player add an extension under its own namespace with no central approval committee. Decentralized governance, then, but it assumes you know what you are declaring. Established

Transports

The specification defines services in REST and JSON-RPC, and supports several transport layers for interoperability. The audit question is not "which should we pick" but "which can our stack expose without a rewrite". That is usually where the gap between a three-week project and a six-month one is decided.

Checkout and its breaking points

A checkout session exposes a status that drives the agent, through incomplete, requires_escalation, ready_for_complete, complete_in_progress, completed and canceled. Errors are graded by severity: recoverable, requires buyer input, requires review, or unrecoverable.

Here the audit maps your real breaking conditions, the ones your teams know and nobody wrote down: age verification, signature thresholds, delivery zones, prescription-only products, pricing exceptions. Each has to find its place inside requires_escalation rather than break the journey. It is the most useful output of phase I, and the one merchants most consistently underestimate.

Payment: instruments and handlers

UCP separates payment instruments from the handlers that execute them. That separation is what lets a merchant change provider without renegotiating its agentic integration. The audit checks whether your current chain supports that distinction, or what it would take. Established

Two framing points save months of pointless committee debate. The Merchant of Record does not change: you remain the seller. And the Checkout capability introduces no PCI DSS compliance obligation for card payments. The question "do we lose control of the transaction" has a written answer, and it is no.

Eligibility and the autonomy dial

UCP provides for eligibility verification: claims must be resolved before a transaction completes. That is the mechanism behind contract pricing, geographic restrictions and customer-specific terms. The audit inventories what you condition today, and whether it is expressible.

Finally, the autonomy dial sits in exactly one place: an order must be finalized manually by the user through a trusted UI, unless the AP2 Mandates extension is supported. A merchant targeting automatic replenishment needs AP2 mandates. A merchant targeting assisted discovery does not. That call, made in phase III, changes the cost of the project threefold. Established

Sources: UCP specification overview and Checkout capability, version 2026-04-08. For implementation, see the implementation guide and the use cases by business type.

Deliverables

What you receive, concretely.

01

Scoping document

PDF of 12-20 pages, structured by layer, claim-tagged, citable internally.

02

Replicable UCP score

Excel/Sheet grid provided with usage instructions to re-run the audit in 12 months.

03

Executive summary

One A3 page or 4 slides, 5 minutes reading, 10 minutes Q&A.

04

Technical partner briefing

Tech brief for vendor / integrator, derived from the roadmap.

Methodological principles

  • Claim tagging. Every statement in the audit is labeled established / emerging / prospective.
  • Platform-neutral. No vendor recommendation is forced. The technical partner is an output, not a prerequisite.
  • Replicable. The scoring grid is provided, with documentation, to repeat the process later.
  • Scope discipline. No piling up of recommendations. Three decisions, not thirty.