AI shopping agents — ChatGPT, Gemini, Perplexity, Claude — don't read your product page the way a human does. They pull from the Shopify Catalog, compare your listing against every other merchant carrying the same SKU and recommend whichever listing matches a shopper's prompt most cleanly. The listing wins, not the product.
The Agentic Product Listing (APL) is the version of your product written specifically for that audience. This guide walks you through creating one, reviewing what AgentIQ generated, fixing your human-facing page based on what was learned and mapping the result back into Shopify so AI agents see it.
What an Agentic Product Listing Actually Is
Your Shopify product has two audiences:
Human shoppers read your product detail page — title, description, images, reviews.
AI shopping agents read the Shopify Catalog — a structured feed Shopify exposes for agentic commerce.
AgentIQ creates a separate, agent-optimized version of each product (the APL) that lives as Shopify metafields. Once you map those metafields into the Shopify Catalog, agents see the optimized version. Your human-facing product page is untouched.
This separation is the point. You can write copy that converts humans without compromising what agents need, and vice versa.
The Workflow at a Glance
There are four things you'll do, in this order:
Review your Catalog Audit in My Store to see which products need work
Create an Agentic Product Listing for a product
Apply AgentIQ's suggestions to your human-facing product page
Map AgentIQ fields to your Shopify Catalog (one-time setup, after your first APL)
Steps 1–3 are per product. Step 4 is once per store.
Before You Begin
You'll need:
AgentIQ installed on your Shopify store
At least one published product in your catalog
Edit access to the Shopify Catalog Mapping screen in your Shopify admin
Step 1: Open Your Catalog Audit
In AgentIQ, click My Store in the left navigation.
This is your Catalog Audit. Every product in your store is scored on Health (Poor, Mediocre, Excellent) based on how well its data lines up with what AI agents look for. The Suggested guardrails column flags specific issues — "Description Too Short", "Over Tagged" — so you know where the gaps are before you start.
Two columns matter for this workflow:
Agentic product listing — products without an APL show a Create APL button. Products with one show View APL.
Product Schema — links to the human-facing product editor where you'll apply suggestions later.
Sort by Health (Poor first) or by Alerts to find the products that need attention. Pick one to start with.
Step 2: Create an Agentic Product Listing
Click Create APL on the product you picked. AgentIQ opens a three-step flow: Prompts → Run → Review.
2a. Pick Your Prompts
Prompts are the shopper queries you want this product to win — the things people actually type to an AI agent. Pick up to three. AgentIQ will search the Shopify Catalog for each prompt, study the top-ranked competitor listings and use what it learns to write yours.
You have three sources:
Suggested — three prompts AgentIQ generated for this specific product based on its category and attributes. A reasonable starting point if you're not sure where to begin.
Library — pull from your existing Prompt Libraries (120 prompts in the example above). Use these when you've already done prompt research and have a curated set.
Manual — type a custom prompt. Use this for queries unique to your brand or audience.
You can mix sources — for instance, one Suggested prompt plus two from your Library. Three prompts is the maximum because more dilutes the signal: the system has to balance optimizing for too many different intents at once.
When you have your prompts selected, click Run agentic listing.
2b. Let It Run
AgentIQ runs through seven steps:
Locating product in Shopify Catalog — matches your product against the public catalog index by title and handle
Reading current agentic listing — snapshots how Shopify currently surfaces this product to AI shoppers
Searching catalog for your prompts — pulls top-ranked competitor listings for each of your three prompts
Fetching product schema — reads your product fields and metafields from Shopify Admin
Reasoning over competitors — extracts what the winning listings have in common and writes your optimized version
Writing APL metafields — saves the new agentic title, category and description to your product as metafields
Verifying catalog re-index — polls Shopify Catalog until the new content is live
Step 7 can take up to three minutes. You don't have to wait — click Run in background, you will see View APL button on the products table when the process in complete.
2c. Review the Result
When the run completes, you land on the APL review page.
This is the agentic version of your product. Read every field:
Unique Selling Point — the single positioning statement an agent will use to differentiate your product
Agentic Title — the title agents will surface (different from your human-facing title)
Agentic Description — the structured description optimized for agent ingestion
Top Features — the bullet points agents will quote when recommending your product
On the left, Top competitors surfaced shows you the actual listings AgentIQ studied for each of your prompts. This is where the recommendations came from — useful context when something looks off.
The green banner at the top is worth re-reading: editing the agentic product listing does not affect your human-facing product page. The APL is what's surfaced to AI shoppers and search agents. Your Shopify product page stays exactly as it was.
If anything needs adjustment, use Edit prompts in the top right to rerun with a different set, or edit fields inline.
Step 3: Apply AgentIQ Suggestions to Your Human-Facing Listing
Creating the APL is half the work. The other half is fixing the human-facing product page based on what AgentIQ learned from your competitors.
Click Product listing from the APL review page (or click the Fix Product button next to that product in My Store).
This is the product editor. At the top you'll see a banner telling you how many AgentIQ suggestions need review, and quick-glance cards for any structural issues — Description Length, Tag Issues, missing alt text.
The suggestions are inline. Each one shows:
Impact level — High Impact or Medium Impact, based on how much the change is likely to move agent ingestion quality
Why it's suggested — a one-line explanation of what's wrong with the current field and what the competitors do differently
The proposed replacement — editable before you apply
You have three options for each suggestion:
Apply — accept the proposed text as-is
Edit — open the suggestion in the inline editor and adjust before applying
Dismiss — reject the suggestion (use this when the suggestion conflicts with your brand voice or merchandising decisions you're confident about)
Once you apply a suggestion, an "AgentIQ Suggestions Applied" indicator appears under the field. Work through every suggestion — title, description, image alt text, metafields — then click Save in the top bar to commit your changes to Shopify.
Unlike the Agentic Product Listing, the Product listing, does affect your visible data on your PDPs, so be mindful of what you are accepting.
The product editor also shows your tracked prompts in the Prompts Analysis panel. Suggestions are scored against these prompts, so you can see at a glance which of your shopper queries each change helps you win.
Step 4: Map AgentIQ Fields to Your Shopify Catalog
This step is what connects everything together. Without it, agents will still read your old Shopify catalog fields instead of the optimized APL versions.
You only need to do this once per store, after creating your first APL.
In Shopify admin, go to Shopify Catalog Mapping.
In the Product fields section, set each standard field's source to the matching AgentIQ value:
Shopify Catalog Field | AgentIQ value |
Product title | AgentIQ Title |
Product description | Agentic description |
Product category | AgentIQ Category |
Save the mapping.
The Product summary panel on the right shows you a live preview — before mapping, the fields display your default Shopify values; after mapping, they'll switch to the APL versions.
If the AgentIQ values don't appear in the source dropdown, double-check that you've completed at least one APL run. The values only exist after a successful run.
What Happens Next
Once mapping is saved, every new APL you create flows through automatically. You don't need to repeat Step 4. From this point on:
AI shopping agents querying the Shopify Catalog see your optimized agentic listings
Your human-facing product pages stay unchanged unless you edit them
New products without an APL still fall back to your default Shopify fields — so creating APLs is how you expand coverage
The goal is to work through your catalog by Health score: start with Poor, then Mediocre, then Excellent. Products in Poor are losing to competitors right now. Products in Excellent are already winning, but an APL still tightens the recommendation.
You can check progress any time from My Store — Health scores update automatically as you create APLs and apply suggestions.