A Shopify product page can display the right information to a shopper while Shopify Catalog receives a different field, an incomplete value, or a title that was never intended to be customer-facing.
This can happen when a store relies on metafields, metaobjects, product tags, title conventions, or custom catalog logic. The storefront may be correct, but the source selected for an AI commerce channel may not be the source your team expects.
Shopify provides a Channel mapping page for Shopify Catalog. It lets merchants choose the source for selected product fields, choose which price is sent as the list price, and configure custom variant grouping. It is optional. Stores using Shopify's standard product fields may not need to change anything.
This guide focuses on when and how to review those mappings without turning a straightforward catalog into an unnecessary data-engineering project.
What Shopify Catalog mapping changes
Shopify Catalog structures product information for discovery across supported agentic storefronts. By default, it uses product information such as title, description, options, images, price, availability, and other key attributes.
The Channel mapping page lets you customize selected parts of that data. Shopify currently documents three relevant areas:
- Product fields: Choose the source for the product title and product description. You can use a product field or metafield, and for text fields you can combine sources with text or use Liquid code.
- List price: Choose whether Shopify Catalog uses the current price or the compare-at price as the list price.
- Custom variant grouping: Group separate Shopify products that are variants of the same product into a single listing.
These settings customize the data Shopify Catalog receives. They do not rewrite the original product structure in your online store, and they do not tell you exactly how every AI channel will display a product.
See Shopify's documentation on customizing product data for Shopify Catalog.
When you should review your mappings
Start with a mapping review when one or more of these conditions is true:
- Your customer-facing product description lives in a metafield rather than Shopify's default description field.
- Your product title contains internal codes, collection labels, supplier names, or delimiter-based variant information.
- Different product types use different metafields or metaobjects to store their main descriptions.
- Separate Shopify products represent color, size, material, or configuration options that should appear as one product listing.
- The price shown to shoppers and the price intended as the list price for AI agents are not the same field or value.
- A catalog preview shows an incomplete, generic, or incorrectly sourced title or description.
Do not change mappings simply because a checklist says every store should customize them. Shopify uses default sources when no custom mapping is selected, and those defaults may already be correct for your store.
Step 1: Inspect a representative set of products
Do not begin by changing the whole catalog. Select a small sample that reflects the way your store is built:
- One standard product using the default Shopify title and description fields
- One product whose main description is stored in a metafield
- One product with multiple variants
- One product with unusual naming conventions or a product tag used for grouping
- One product with a current sale price and compare-at price
- One product that has recently been added or changed
For each sample, record the information a shopper should see and the field where that information currently lives.
| Product information | Intended source | What to verify |
|---|---|---|
| Product title | Shopify product title or a title metafield | Human-readable name, without internal codes or accidental suffixes |
| Main description | Shopify description or a description metafield | Accurate product purpose, differentiating details, and limitations |
| Current price | Shopify price | Matches the amount a customer can actually pay for the selected option |
| Compare-at price | Shopify compare-at price, where legitimately used | Represents a real reference price under the store's pricing policy |
| Variant options | Shopify product options or mapped grouping logic | Color, size, material, or configuration values are explicit and consistent |
The point of this first pass is to establish the intended source of truth before you touch Channel mapping.
Step 2: Open the Shopify Catalog mapping page
In Shopify Admin:
- Go to Sales channels and open Agentic.
- In the Sources section, expand Shopify Catalog.
- Select Review product data mapping.
- Confirm that Shopify Catalog, rather than another sales channel, is selected on the Channel mapping page.
Shopify's help documentation describes this path as the way to review the fields used by Shopify Catalog. The exact labels in the admin can change, so use the current Shopify instructions if your interface differs.
Step 3: Review title and description sources
In the Product fields section, review the existing source for the product title and product description.
For each field, ask:
- Is this the field customers would recognize as the product's real name or description?
- Does it contain complete, factual information?
- Does it include internal data that belongs in operations rather than product discovery?
- Is the value populated for all relevant products?
- Does the value change by product category or product type?
Example: the real description is stored in a metafield
Imagine a store uses Shopify's default description field for a short teaser, while a metafield named Full description contains the detailed, customer-facing product information.
The storefront template may render Full description, but Shopify Catalog may still be mapped to the default description field. In that situation, review whether mapping the Catalog description to Full description would provide a more accurate representation.
Before saving, check that the metafield is populated for the products that use it. Shopify documents that when a selected mapping source has no value for a product, the field's default source is used instead. A missing metafield value therefore does not automatically create a complete catalog record; it may cause different products to fall back to different sources.
Use field coverage as a warning signal
Shopify displays a field-coverage percentage for each mapped field. It represents the share of products with a value for that field. Review the products behind a low coverage percentage and decide whether the missing values are intentional or need to be fixed.
Do not aim for 100% merely for the sake of a perfect dashboard. A value should only be populated when it is accurate and relevant. A blank field is safer than a fabricated specification.
One current limitation matters: Shopify's documented mapping options allow you to change the product title and description sources, but do not let you map the product category for Shopify Catalog.
Step 4: Decide which price should be the list price
Shopify Catalog lets you choose between:
- Price: the product's current price is used as the list price.
- Compare-at price: the compare-at price is used as the list price when it is higher than the current price.
AI agents may use a list price to show a strikethrough price for a discounted product. This makes the setting a merchandising decision, not just a technical preference.
Before changing it, review:
- Whether your compare-at prices are accurate and maintained consistently.
- Whether the selected option should represent the regular reference price or the current selling price.
- Whether the product page, catalog data, and checkout show a consistent promotion.
- Whether variant prices differ and need to be checked individually.
Do not use a compare-at price simply to make a discount appear larger. The reference price should follow the store's actual pricing practices and applicable consumer-protection rules.
Shopify notes that an eligible compare-at price can be sent through the list_price field in the Universal Commerce Protocol Variant specification. This is a data representation detail, not a guarantee that every AI channel will render a discount label in the same way.
Step 5: Preview more than one product
Use the product preview to inspect the result of a mapping change. Check the products with the most complicated data structures, not only the product that motivated the change.
A useful review sheet looks like this:
| Sample | Expected title | Expected description source | Price behavior | Result |
|---|---|---|---|---|
| Standard item | Default title | Default description | Current price | Pass / review |
| Metafield item | Default title or title metafield | Full description |
Current price | Pass / review |
| Sale item | Correct customer-facing title | Correct description source | Compare-at or current price, as intended | Pass / review |
| Multi-option item | Clear parent name | Shared or variant-specific content as appropriate | Correct by selected variant | Pass / review |
There is an important limitation: Shopify says the product preview shows how Shopify Catalog receives the data. It is not an exact indicator of how AI channels will display or rank the item.
Treat the preview as a mapping check, not as a recommendation simulator.
Step 6: Save, allow for processing, then retest
Shopify states that mapping changes are processed with a delay and may not immediately affect the data used by Shopify Catalog.
Record:
- The products you sampled
- The field sources before and after the change
- The date and time of the change
- The field-coverage percentage
- The products that still need manual correction
- The date and time of the follow-up check
After the change has had time to process, review the product data again. Then run a small, repeatable prompt set for the affected products. Record the exact prompt, market, date, and whether the product appears. A single answer is not proof that the mapping change caused a recommendation.
For a more complete diagnostic process, see How to Audit a Product Catalog for AI Shopping and AI Shopping Readiness Scores: What They Measure and What They Don't.
Common mapping mistakes
Mapping every product to one metafield
A metafield may be complete for one product type and empty for another. Use field coverage and representative product checks before applying a global mapping.
Treating a preview as a channel ranking result
The preview confirms the data Shopify Catalog receives. It does not reproduce the full selection or presentation behavior of ChatGPT, Google AI Mode, Gemini, or another channel.
Forgetting that a mapping can have a fallback
When a selected source is empty for an item, Shopify may use the default source. That can create inconsistent descriptions across a catalog if the team does not review coverage.
Changing list price without reviewing promotion logic
A compare-at price should be a genuine reference price. Confirm the selected behavior across products and variants before using it as the Catalog list price.
Changing too many fields at once
When several fields and grouping rules are changed simultaneously, it becomes difficult to identify which change caused an improvement or a regression. Change one logical area at a time and keep a change log.
What field mapping can and cannot do
A good mapping can help ensure Shopify Catalog receives the product information your store intends to share. It can reduce the risk that the channel receives an internal title, an abbreviated description, or the wrong list-price source.
It cannot guarantee that a product will be shown for a particular prompt, reveal private recommendation weights, or make an unsuitable product relevant to a shopper's needs.
The goal is not to manipulate an AI ranking system. It is to make the product data supplied by your store accurate, coherent, and useful.