Skip to content 
Most WooCommerce store owners assume their product catalog is in good shape. The products are listed. The prices are correct. The images look fine.
Then they run an AI agent readiness audit and find out that a significant portion of their catalog is invisible to every AI system that matters: ChatGPT, Perplexity, Google AI Mode, and the AI agents now completing purchases on behalf of customers.
The problem is not that the products do not exist. It is that the data around those products is incomplete or formatted in a way that humans can read but machines cannot parse with confidence.
AI agents evaluate structured signals and when those signals are missing or ambiguous, they move on to a competitor whose catalog gives them a cleaner answer.
This audit walks you through exactly what to check, area by area, so you know where your catalog stands and what needs fixing.
If you are new to the topic, our introduction to agentic commerce for WooCommerce covers the broader picture first.
Key Takeaways
- According to a March 2026 BigCommerce survey, 40 percent of ecommerce businesses are still standardising product pages for agentic AI, and 33 percent have not started at all. The first-mover opportunity is wide open.
- Velou’s research puts it plainly: an 80 percent complete catalog is not 80 percent as effective. Missing fields make products functionally invisible for any query requiring those fields.
- The audit covers six dimensions: product data completeness, schema, catalog accessibility, pricing and inventory accuracy, policy machine-readability, and image quality.
- Some fixes require no developer involvement. Others do. This audit helps you separate the two clearly.
- The same improvements that make your catalog readable by AI agents also improve Google rich results and your overall conversion rate.
Why Your Catalog Needs a Dedicated AI Audit
A standard SEO audit and an AI agent readiness audit are not the same thing.
Traditional SEO audits focus on crawlability, keywords, and backlinks.
AI agents evaluate your catalog on a different set of criteria: data completeness, structural consistency, and the presence of specific fields that let them make confident recommendations and complete transactions without human intervention.
A product page that ranks on page one of Google can still be invisible to an AI agent if it is missing a GTIN, has no aggregate rating in its schema, or describes its return policy only in paragraph text. The audit below is designed specifically to surface those gaps.
Q: Can’t I just use my existing SEO audit results? No. SEO audits measure crawlability, keyword coverage, and link signals.
AI agents evaluate structured data fields, schema completeness, and machine-readable policies.
A page that passes an SEO audit with flying colours can still be completely invisible to an AI agent purchasing on a customer’s behalf.
Area 1: Product Data Completeness
This is the foundation. Before structured data, before schema, before anything else, the raw product data in your WooCommerce catalog needs to be complete.
AI agents evaluate products by parsing available data fields and matching them against query requirements. When a field is empty, the agent has no signal to work with for that dimension.
When too many fields are empty, the agent loses confidence in the product overall and is less likely to recommend or purchase it.
What to Check
Go through your top 50 SKUs (or your full catalog if it is manageable) and assess each product against this list.
Product Title Clarity
Product title clarity: Does the title describe the product in the way a buyer would naturally ask about it, including brand, key attribute, and product type? A title like “Blue Widget Pro 500ml” gives an agent three clear parsing signals. A title like “Our Best Seller” gives it none.
Product Description
Product description length and specificity: Descriptions under 300 words tend to perform poorly in AI citations. Check whether your descriptions answer the questions a buyer would actually ask an AI assistant, rather than using internal or manufacturer vocabulary.
Brand Field
Brand field: Is the brand explicitly populated for every product as a separate WooCommerce field? AI agents use brand as a primary matching signal for queries like “best [brand] supplement for energy.”
SKU
SKU: Every product and every variant should have a unique SKU. Missing SKUs create ambiguity in agent-processed orders and in catalog feed matching.
GTIN (Global Trade Item Number)
GTIN (Global Trade Item Number): This is the field that most WooCommerce stores are missing and that has the highest impact on AI agent visibility.
GTIN includes EAN, UPC, and ISBN depending on product type.
Products without a GTIN are harder for agents to identify with confidence, especially when comparing across multiple stores.
Attributes and Variants
Attributes and variants: Are all product variants fully populated with their own data rather than vague descriptors?
An agent selecting the correct variant needs to do so without ambiguity. Gaps cause agent errors or abandoned transactions.
Quick Action
Export your catalog to CSV and filter for empty cells across brand, GTIN, and attribute columns. The number of gaps is your starting count.
Q: Does every product really need a GTIN? For most physical products, yes. GTIN is how AI agents cross-reference your product against external databases to confirm identity, pricing, and availability.
Without it, agents have lower confidence in your product data and are more likely to recommend a competitor who has it.
Products where GTIN genuinely does not apply (custom or handmade items) are the exception, not the rule.
Area 2: Schema and Structured Data
This is where most WooCommerce stores have their biggest gap.
WooCommerce generates basic product schema by default: name, price, and sometimes availability. It omits the fields that AI systems specifically look for. According to research from Google I/O 2026 cited by XICTRON, schema-compliant pages are cited up to 3.1 times more frequently in AI answers than pages without it.
A separate 5,000-site audit by Digital Applied confirmed that sites passing the Rich Results Test cleanly are cited measurably more often across Google AI Overviews, Perplexity, and ChatGPT. That is the difference between being in the recommendation set and not being in it.
What to Check
Run your top product pages through Google’s Rich Results Test. For each page, check whether the schema output includes all of the following:
Name and Description
Name and description: Basic, but confirm they are present and populated from actual product data rather than defaulting to the page title.
Price and Currency
Price and currency: Must be present and accurate. AI agents use price as a primary filter for many purchase decisions.
Availability
Availability: Should use standardised schema values (InStock, OutOfStock, PreOrder) rather than free-form text like “usually ships in 3 days.” Free-form text is human-readable. Standardised enum values are machine-readable.
Brand
Brand: As a structured schema field, not just mentioned in the description.
GTIN in Schema
GTIN: In the schema output, not just in the product admin. Many WooCommerce setups store the GTIN in a custom field but do not surface it in the schema.
AggregateRating
AggregateRating: If your store has product reviews, the aggregate rating and review count should appear in your schema. This is a trust signal that AI agents weigh when choosing between competing products.
MerchantReturnPolicy
MerchantReturnPolicy: This is one of the most commonly missing schema fields in WooCommerce stores and one of the most actively used by AI agents evaluating purchase confidence. If your return policy is only described in paragraph text on a standalone page, it is not machine-readable.
OfferShippingDetails
OfferShippingDetails: Shipping timeframes and costs in structured schema format, not just on a policies page.
Quick Action
Run three of your best-selling product pages through Google’s Rich Results Test. Note every field that shows as missing or invalid. That list is your schema priority queue.
Q: My store uses Yoast or Rank Math. Does that mean my schema is already complete?
These plugins significantly improve on WooCommerce’s default schema output, but they do not automatically populate every field AI agents look for.
GTIN, MerchantReturnPolicy, OfferShippingDetails, and variant-level structured data typically require additional configuration beyond default plugin settings.
Running the Rich Results Test is the only way to confirm what is actually being output.
Area 3: Catalog Accessibility to AI Crawlers
A perfectly structured catalog is worthless if AI crawlers cannot access it. AI systems send their own crawlers with distinct user agent names. If your robots.txt blocks them, no AI system can read your product pages regardless of how well optimised they are.
What to Check
Access your robots.txt file at yourdomain.com/robots.txt and check for any rules that explicitly block or could inadvertently block: GPTBot, OAI-SearchBot, ClaudeBot, and PerplexityBot.
Also check for wildcard disallow rules that block all crawlers except a specific whitelist. These are common in older WooCommerce installations and security plugin configurations, and they catch every AI crawler automatically.
Quick Action
Check robots.txt for the four crawler names above. If any are blocked, removing those blocks is the fastest and most impactful single fix in this entire audit.
Q: Will unblocking AI crawlers cause any security or performance issues? No. AI crawlers behave like standard web crawlers. They read publicly available pages and do not interact with your store’s checkout or admin. The only practical consideration is crawl rate. If your hosting is on a very constrained plan, you can set a crawl-delay directive rather than blocking entirely.
Area 4: Pricing and Inventory Accuracy
This area is specifically critical for agentic commerce and is often overlooked in content-focused audits.
AI agents completing purchases on behalf of customers need to trust that the price in your catalog is the price they will actually be charged, and that the product is available when they submit the order.
Stale or inconsistent signals cause failed transactions and reduce the likelihood of your store being used in future agent purchases.
What to Check
Price Consistency
Price consistency: Is the price on your product page identical to the price in your product schema and any marketplace feeds? Inconsistencies can cause agent transactions to fail at the payment stage.
Sale Price Accuracy
Sale price accuracy: Is the sale price reflected in the schema Offer block with a valid PriceValidUntil date? An agent that sees a sale price in schema but hits the regular price at checkout will log a mismatch.
Inventory Accuracy
Inventory accuracy: Are out-of-stock products correctly marked as OutOfStock in both WooCommerce and their schema availability field?
Feed Update Frequency
Feed update frequency: How frequently is your product feed refreshed?
Stale feeds are one of the most common sources of price and availability mismatches. Daily updates are the minimum for active catalogs.
According to the Digital Applied AI Agent Readiness Framework, real-time inventory webhooks are the standard for full agentic commerce readiness.
Quick Action
Pick five products that have been on sale recently. Check whether the sale end date was correctly handled in both WooCommerce and the schema output.
Area 5: Policy Machine-Readability
Return policies and shipping policies are trust infrastructure. Human buyers read them occasionally. AI agents evaluate them for every transaction they consider on a customer’s behalf.
If that information exists only as paragraph text on a standalone page, the agent either cannot find it reliably or cannot parse it with confidence.
What to Check
Return Policy Schema
Return policy schema: Is your return policy expressed as a MerchantReturnPolicy schema block, with return window and return method in structured fields?
This is developer work but among the highest-impact fixes for agentic commerce readiness.
Shipping Policy Schema
Shipping policy schema: Is estimated delivery time and cost in OfferShippingDetails schema on your product pages, or only in prose on a policies page?
Policy Language Clarity
Policy language clarity: “Returns accepted within 30 days of delivery for unused items in original packaging” is clear and parseable. “We want you to be happy with your purchase” is not.
Quick Action
Check whether your product schema includes a MerchantReturnPolicy block. In most WooCommerce stores it is absent. Its absence is fixable with developer support.
Area 6: Image and Media Quality
Alt text on product images is a structured signal that AI crawlers use to understand what the image depicts.
Missing or generic alt text (“image1.jpg”, “product photo”) provides no signal. Descriptive alt text that includes product name, key attributes, and brand gives AI systems an additional confidence layer.
AI shopping assistants also use image recognition to match products across sources.
A product with high-quality images on a clean background is more reliably matched than one with low-resolution or cluttered photography.
What to Check
Alt Text Coverage
Alt text coverage: Check the alt text field for your top 20 product images in the WooCommerce media library. How many are blank or generic?
Image Resolution
Image resolution: Do your primary product images meet the minimum standard for AI shopping platforms (typically 800×800 pixels, clean background)?
Image Count per Product
Image count per product: Products with a single image give AI systems less to work with than products with multiple angles and variant-specific images.
Quick Action
Check alt text coverage across your top product images. Blank alt text is a fix that requires no developer involvement and can be done systematically.
Scoring Your Catalog
After working through the six areas, take your top 50 SKUs and score each one against the key fields from Areas 1 and 2 (title, brand, GTIN, description length, schema completeness, return policy schema, availability). Each populated field scores a point.
The percentage is your readiness score.
Most WooCommerce stores score 30 to 55 on the first pass.
That reflects the fact that AI agent readiness standards are newer than most catalog configurations, not a criticism of how the stores are built.
The practical thresholds, drawn from the Digital Applied readiness framework: under 60 is urgent, 60 to 80 needs improvement, 80 and above is agent-ready.
What to Fix Yourself vs. What to Hand to a Developer
Not everything in this audit requires developer involvement. Being clear about the distinction saves time and helps you prioritise.
You Can Fix Without a Developer
- Product titles and descriptions
- Brand field population
- Alt text on product images
- Robots.txt crawler blocks
- SKU consistency across variants
Requires Developer or Plugin Configuration
- GTIN field implementation and schema surfacing
- MerchantReturnPolicy and OfferShippingDetails schema blocks
- Availability enum standardisation (InStock vs free text)
- Sale price PriceValidUntil fields in schema
- Real-time feed update infrastructure
A focused engagement on schema completion, policy structured data, and feed infrastructure has a defined scope and a measurable outcome.
Frequently Asked Questions
Do I Need to Audit Every Product, or Just My Best Sellers?
Start with your top 50 SKUs by revenue. These are the products most likely to be queried by AI agents. Once the framework is established, applying the same standards to the rest of the catalog is faster.
How Often Should I Re-Audit My Catalog?
Quarterly. Plugin updates, theme changes, and WooCommerce core updates can all introduce schema regressions without any visible change to your store. What passes the Rich Results Test today may fail after an update.
Does Fixing My Catalog Help with Regular Google SEO Too?
Yes, directly. The schema completeness and product data improvements in this audit are the same changes that unlock Google rich results (price, rating, and availability badges in search listings). According to WooCommerce’s own 2026 guide on AI-driven commerce, the work that makes your store visible to AI agents also makes it better for human shoppers. The audit is a single investment with benefits in both directions.
What Is the Biggest Single Fix for Most WooCommerce Stores?
Robots.txt cleanup followed by GTIN implementation. Unblocking AI crawlers takes minutes and has an immediate effect. Adding GTINs to your top products addresses the most common data gap that prevents agent-level product matching.
How Does This Connect to the Server-Side Tracking Work?
Directly. Clean, complete product data in your catalog is what makes server-side tracking meaningful.
When an agent order fires a server-side purchase event, that event carries your product data as the payload. If the catalog data is incomplete or inconsistent, the tracking data is too.
The two workstreams reinforce each other. Our server-side tracking post covers the tracking side of that picture in detail.
The Bigger Picture
The catalog audit is unglamorous work. It is systematically going through data fields, checking schema outputs, and fixing gaps no human visitor would ever notice.
But it is the highest-leverage technical investment a WooCommerce store can make for agentic commerce readiness right now.
Server-side tracking captures orders accurately only if the product data is clean. AI search visibility improves fastest when schema is complete. Agentic purchase flows succeed only when pricing, availability, and policy data are structured and accurate.
Everything in the agentic commerce stack depends on the catalog being right.
If the audit above has surfaced more fixes than your team has bandwidth to handle, CoSpark’s WooCommerce team works on exactly this kind of catalog infrastructure work.
