What is agent readiness for a fashion catalogue?
Agent readiness is the measurable condition in which a catalogue gives a shopping agent enough structured, current and unambiguous evidence to match products against a user’s constraints. It covers descriptive attributes, product identity, variant structure, price and availability, machine-readable publication and freshness.
It does not mean that a merchant can guarantee inclusion or ranking. The shopping channel controls retrieval and recommendation. The merchant controls whether the catalogue supplies reliable evidence.
Why can an AI shopper ignore an otherwise suitable product?
Consider the request: “Find a sustainable black blazer under €200, with no polyester, available in my size and deliverable by Friday.” A product can be a genuine match and still fail the machine comparison if its colour is buried in an image, material is written as “premium blend”, variants have no named size or stable identifier, price or availability is stale, or delivery data is absent.
| User constraint | Evidence the catalogue needs | Typical failure |
|---|---|---|
| Black blazer | Category, colour and product type | Colour exists only in imagery or inconsistent tags |
| Under €200 | Current price and currency | Feed and storefront disagree |
| No polyester | Explicit fibre composition | “Sustainable fabric” without composition |
| My size | Named variants and availability | Default option names or stale stock |
| Arrives Friday | Eligible market, fulfilment and delivery data | No machine-readable delivery promise |
Which catalogue fields matter first?
For fashion, the highest-value foundation is usually: a clear title and description; standard product category; colour, material composition, care, gender or audience and origin where relevant; SKU and valid GTIN where available; named size and colour variants; usable imagery and alt text; current price and availability; and consistent product URLs and structured data.
Google recommends unique identifiers for variants and explicit variant properties such as size, colour and material. Shopify says its Catalog syndicates title, description, options, images, price, availability and other structured attributes to eligible AI channels. Those are channel facts, not a promise that completing a field causes a ranking.
What does Cycle diagnose today?
Cycle’s Shopify Agent Readiness scan is read-only and evaluates four explainable layers: product data, machine-readable identity, estimated channel exposure, and freshness and consistency. It weights findings with recent unit sales so a gap on a commercially relevant product is not treated like a gap on a dormant item.
The current scan identifies missing filterable attributes, standard category, GTIN and sufficiently descriptive copy. Its channel-exposure score is explicitly provisional until live storefront and protocol verification is implemented.
What can Cycle correct today, and what remains direction?
Available now: the connected Shopify app can diagnose the catalogue and perform selected reversible changes — including image alt-text completion and normalization of supported catalogue values — through preview, confirmation, logging and undo. Not yet a universal bulk fix: category taxonomy, GTIN, attribute and description corrections still require additional controlled publishers. Cycle does not claim to improve a merchant’s ranking; it improves the quality and governability of the product evidence that channels can read.