// SERVICE
FIND OUT WHAT AN AI SHOPPING AGENT ACTUALLY SEES WHEN IT READS YOUR CATALOG
A 5-day audit that grades your catalog against the real feed specs, checks whether agent crawlers can reach you at all, and returns a per-surface eligibility matrix for ChatGPT, Copilot, Gemini, and Perplexity.
// ENGAGEMENT
one-time build
// TIMELINE
5 business days
A graded eligibility report across four AI shopping surfaces, plus a prioritized fix list.
Fixed
Fixed Scope
5 days
Delivery
4
Surfaces Assessed
100%
Credited Toward the Sprint
// FEATURES
WHAT'S INCLUDED
Feed-Field Coverage, Product by Product
We grade your catalog against the OpenAI product feed spec — item_id, gtin/mpn, brand, availability, seller_url, and the is_eligible_search / is_eligible_checkout flags. Missing GTINs and non-unique IDs are two of the most common reasons products get excluded outright.
Agent Crawler Access Check
Bot-protection and WAF rules block AI search crawlers more often than robots.txt does, a layer below where anyone thinks to look. We test OAI-SearchBot, PerplexityBot, and the rest from a non-whitelisted IP, and separate the search crawlers from the training crawlers so you can make that call deliberately rather than by accident.
Price + Availability Parity
Feed says in stock, the PDP says backordered — or the two prices simply don't match. Mismatches between your feed and your live page are a named rejection reason and an easy way to lose eligibility silently. We diff them across a sample of your catalog.
Per-Surface Eligibility Matrix
ChatGPT, Microsoft Copilot, Google's Gemini surfaces, and Perplexity each have different requirements and different levels of Shopify automation. You get a grid showing where you're eligible, where you're enrolled, and where you're neither.
PDP Extractability + Schema Review
Not every agent path runs through a feed — some read the page. We check that your Product and Offer schema is present and accurate, and that price, availability, and policy content survive in the server-rendered HTML rather than appearing only after JavaScript runs.
A Fix List You Can Act On Without Us
Every finding is scoped, prioritized by impact, and priced. If you have an in-house developer, hand them the document. The audit is useful whether or not you hire us for the implementation.
// PROCESS
HOW IT WORKS
// STEP 01
Pay online — no scoping call required to start
// STEP 02
You grant read access to your store admin (or export the catalog)
// STEP 03
We pull your catalog, crawl your PDPs, and test each surface
// STEP 04
Findings are reviewed by hand — no raw tool output gets shipped
// STEP 05
PDF delivered in 5 business days, with a walkthrough call if you want one
// FAQ
COMMON QUESTIONS
// RELATED