Google Conversational Attributes
Enrich your Google Merchant Center feed with Conversational Attributes to support Google's AI-powered shopping experiences.
AI-driven shopping is changing how buyers find and evaluate products — through AI Mode in Search, conversational agents, and other AI-powered surfaces. To help these systems represent your products accurately, Google Merchant Center supports a set of conversational attributes.
These attributes let you submit richer, more nuanced product information than a standard feed allows: answers to the questions buyers actually ask, links to related products and accessories, and detailed variant options such as size, color, and fit. Feeding this context to Google helps AI systems and conversational agents understand your products the way a well-briefed salesperson would, so buyers find the right product faster — whether they're browsing traditional search results or asking an AI assistant a question.
This feature is built for Productsup clients who already export a product feed to Google Merchant Center and want to enrich that feed with conversational depth, without touching or risking the integrity of their core product data.
Available conversational attributes include:
| Attribute | What it's for | Example |
|---|---|---|
Question and answer question_and_answer | Product-specific FAQs — the questions buyers actually ask, and your answers | "Does it have a headphone jack?" → "This version doesn't have a headphone jack." |
Document link document_link | Links to related PDFs, such as manuals or assembly instructions | https://example.com/manual.pdf |
Related product related_product | Cross-sells, accessories, or required parts, linked by identifier | accessory:gtin:811571013579 |
Item group title item_group_title | A shared title for a product with multiple variants | "Google Pixel 9" |
Variant option variant_option | The specific properties that distinguish one variant from another | "display:XL,memory:512GB,color:moonstone" |
Popularity rank popularity_rank | How a product ranks in popularity as a percentage of your inventory | 95.5 |
These attributes are entirely optional and don't affect the approval status of your existing products. If you already populate detailed information in your description, product_highlight, or product_detail attributes, you don't need to duplicate that same content here — conversational attributes complement your existing product data rather than repeat it.
This data is sent to Google Merchant Center to enrich your main product feed. Before setting up conversational attributes, make sure your primary product feed is already flowing into GMC through one of the supported Productsup export options.
Prerequisites
Google recommends submitting conversational data as a supplemental data source, layered on top of your existing primary feed, rather than folding it into the primary feed itself.
Before setting up the export in Productsup, prepare a supplemental data source in Google Merchant Center:
In your Merchant Center account, go to Settings → Data sources.
Select the Supplemental sources tab, then Add supplemental product data.
The Supplemental sources tab only appears once the Advanced data source management add-on is enabled on your account.
Choose your source type (file, Google Sheets template, or API), and enter:
- File name — must exactly match the file you submit, including its extension.
- Product ID — ties your supplemental data back to the matching product in your primary feed.
- Data source label — a name to help you identify this data in Ads campaigns.
- Language — the language your product data is written in.
Link the new supplemental source to the primary data source it should enrich.
See Google's guide to creating a supplemental data source for full details, and the conversational attributes reference for how each attribute works and how to format its values.
Add the Google Merchant Center Conversational Attributes export
Go to Exports in your site's main menu and select Add export. Search for Google Merchant Center Conversational Attributes, hover over it, and select Add.
Select the newly added Google Merchant Center Conversational Attributes export, then select Add Destination.
Choose one of two ways to deliver conversational data to GMC, matching how your supplemental data source is configured:
- Productsup Server — hosts the generated file on Productsup's servers. Provide the file's URL as the source when configuring the supplemental data source in GMC, or upload it manually. See Google's guide on uploading products to Merchant Center for how GMC consumes a hosted file.
- SFTP/FTP/FTPS — use the credentials Google provides when you set up the supplemental feed data source in GMC.
See Choose how you want to send data to GMC for the full walkthrough of both destination options.
Go to Dataflow to locate the conversational attributes available for Google Merchant Center, and map them from import to export.
Caution
Use the same product identifier you use in your GMC main feed. Google matches conversational data to existing products by ID — if the identifiers don't align, your enrichment data doesn't attach to the right products and the update has no effect.
Go to Data View to review and optimize the data before it goes live. Use this step to catch formatting issues, verify Q&A pairs, and confirm variant options are structured correctly.
Best practices
- Match identifiers exactly. The product ID in your conversational attributes feed must mirror the ID in your GMC primary feed, character for character — this is the most common point of failure.
- Don't duplicate what you already have. If your
description,product_highlight, orproduct_detailattributes already cover a piece of information, leave it out of the conversational attributes to avoid redundant, bloated feeds. - Start with your best Q&A content. Pull from real customer questions — support tickets, live chat logs, product review comments — rather than guessing what buyers might ask. This is where conversational attributes add the most value for AI-driven surfaces.
- Keep variant data consistent. Use the same naming conventions for
variant_optionvalues (for example, alwayscolor, not sometimescolour) across your whole catalog so AI systems can reliably group and compare variants. - Confirm the Advanced data source management add-on is enabled in GMC before you start. Without it, the Supplemental sources tab doesn't appear, and setup stalls at the first step.
- Test in Data View before going live. A quick pass here catches malformed Q&A pairs or missing related-product identifiers before they reach Google.
- Remember this is additive, not destructive. Conversational attributes can't add or remove products, and submitting them doesn't affect your existing products' approval status, making this a low-risk way to enrich what's already working.
How is this guide?