A practical PDP audit for AI search readiness: align visible product facts, Product JSON-LD, Merchant Center feed fields, images, localization, and measurement without promising guaranteed rich results.
AI search does not remove the need for classic product page SEO. It makes the weak spots easier to expose. A product page still needs a clear title, crawlable content, a fast mobile experience, and a canonical URL. But ecommerce teams now also need to keep the visible page, Product JSON-LD, variant attributes, images, policies, and Merchant Center feed aligned.
This checklist is built for DTC teams that already have product pages but want them to be easier for Google Search, Google Shopping, AI Mode-style shopping experiences, and other product discovery systems to understand. The goal is not to promise ranking or rich results. Google explicitly says structured data features are not guaranteed. The goal is to remove avoidable ambiguity so crawlers, feeds, and reporting tools can read the same product facts.
Google's Product structured data documentation separates product snippets from merchant listings. Product snippets can show details such as ratings, price, and availability. Merchant listings apply to pages where shoppers can buy from you and support more product-specific information such as shipping, sizing, returns, and variants. Google also recommends combining page-level structured data with Merchant Center feeds because the two sources help Google understand and verify product data.
AI shopping raises the bar for attribute completeness. Google's May 20, 2025 AI Mode shopping update says AI Mode combines Gemini capabilities with the Shopping Graph, which includes product listings with details like reviews, prices, colors, and availability. On May 27, 2026, Google Merchant Center announced upcoming AI performance insights, including product attribute insights and an attribute completeness score for missing structured product attributes.
That does not mean every Product schema field is an AI ranking factor. It means PDP SEO now has to be operated as a data-quality workflow, not only a copywriting workflow.
Before touching schema, verify the basics:
If these basics are wrong, structured data only makes the inconsistency easier to detect.
The Product JSON-LD on a PDP should describe what shoppers can see on that same page. Audit the core fields first:
Do not mark up hidden claims, fake reviews, or business policies that are not represented on the page. Google's structured data guidelines require structured data to be a true representation of the page content and warn that misleading markup can prevent rich results from showing.
Product schema is not one giant checkbox. Google distinguishes product snippets, merchant listings, variants, shipping, returns, loyalty, and policies. A PDP audit should therefore ask two questions:
For example, price and availability may appear in both page markup and the feed. Shipping and returns may come from Merchant Center, product-level merchant listing markup, or organization-level policy markup. Google documents an order of precedence for shipping and return configurations, so teams should avoid maintaining conflicting values in multiple places.
AI search and shopping surfaces are sensitive to stale product facts. Check these fields across PDP, feed, and backend records:
The practical rule is simple: the shopper-facing page, schema, and feed should not disagree about what the product is, whether it is available, or what it costs.
Attribute completeness matters because users ask product questions in natural language. Instead of only optimizing a title for one keyword, audit whether the page can answer common buying filters:
The May 27, 2026 Merchant Center AI insights announcement is a useful signal here: Google specifically calls out product attribute insights and an attribute completeness score for products missing structured attributes.
Product images should not be treated as decoration. Check that:
Google's structured data guidelines state that image URLs used in structured data must be crawlable and indexable. If the image cannot be accessed, it cannot support richer product presentation.
After deployment, use reporting rather than assumptions:
Treat this as an operating loop: fix invalid items, inspect live URLs, request validation, and compare performance after template or feed changes.
Foundax should be described carefully. It does not guarantee rich results, AI visibility, or automatic product discovery. What it does provide is an operating workflow for keeping product facts consistent:
This is the right framing for AI search: not magic visibility, but fewer data mismatches and a clearer path from product record to public page to merchant feed.
Use this sequence when reviewing a high-priority product page:
No. Product schema can help search systems understand page facts and may make pages eligible for richer Google Search experiences, but Google does not guarantee that structured data features will appear. Treat schema as a data-quality foundation, not a ranking promise.
Start with title, description, images, SKU or identifiers, brand, price, currency, availability, canonical URL, and variant attributes. These fields are the easiest to compare across the visible page, Product JSON-LD, and Merchant Center feed.
Product snippets are product-related search results that can include information such as ratings, price, and availability. Merchant listings apply to pages where customers can buy from you and support more commerce-specific details such as shipping, returns, sizing, and variants.
Only if the page actually has shopper-facing FAQ content and your implementation supports that markup. Do not add FAQ markup just because a checklist says so. Misleading or invisible structured data is a quality risk.
Recheck after template changes, feed changes, pricing or availability automation changes, major localization updates, and any Search Console or Merchant Center warning. High-volume SKUs should be reviewed on a recurring cadence.
Foundax helps teams keep product records, PDP metadata, Product JSON-LD, sitemap/Google workflows, and GMC preflight/sync in one operating path. That reduces mismatch risk across page, schema, and feed, while still leaving business facts under merchant control.
For the broader discovery model, read Product Data Is Becoming the SEO Layer for AI Commerce Discovery, then use the agentic commerce product data guide to prioritize fields across catalog, feed, and storefront.