Google UCP and Why Clean Product Data Matters Now

Think about the last time you bought something online. You probably searched, clicked through a few product pages, compared a couple of options, added something to your cart and checked out. That’s been the way we’ve all done it for about twenty years now, and it’s about to get a lot easier.

Google’s been rolling out something called the Universal Commerce Protocol, or UCP, and it’s going to change how we all find and buy products. As an eCommerce store owner, it’s worth understanding how it works and how you can make sure your store is ready for it. Now, here’s the part most store owners are going to miss: success is going to some down to something more important than your store’s platform – your product data.

What is UCP?

In simple terms, UCP is a common language that lets AI shopping assistants talk directly to online stores. Google describes it as an “open standard designed for the future of commerce,” and the first thing it powers is being able to purchase something using AI Mode in Google Search and Gemini.

Now, this isn’t something new.  Google first announced UCP back in January and Shopify, Etsy, Wayfair, Target and Walmart all helped develop it. It’s open source, so it’s not exclusive to Google, and it works alongside other AI standards like the Model Context Protocol (MCP) and Google’s Agent Payments Protocol (AP2).

What’s new today (yes, right at the start of the holiday shopping season) is that Google is gradually rolling out a UCP hub inside Merchant Center for US merchants. It’s an early access program right now, so not every store can turn it on today, but it’s important to get familiar with it so when it’s available for your store, you’re ready.

Here’s a few key things worth knowing:

  • You stay the seller. You remain the Merchant of Record, and you keep your customer data, your relationships and the post purchase experience.
  • It runs on your existing Google feed. It uses your current Merchant Center shopping feeds to reach shoppers during discovery, so there’s no separate feed needed.
  • It’s basic right now, but more is coming. Google’s roadmap includes multi item carts, linking loyalty accounts and post purchase support for things like tracking and returns.

How This Changes the Way People Shop

Here’s how online shopping is going to change in a big way. You just went to your refrigerator and while it was dispensing ice for your morning Diet Coke, the “change filter” light lit up.  Pull your phone out of your pocket, and tell Google or Gemini “I need a replacement filter for my Whirlpool fridge, model WRS325SDHZ, delivered by Friday.” AI does the searching, narrows it down to the right product, and all you do is hit Buy now right there in the conversation without ever loading the store’s website.

gemini screenshot of filter results

That’s a huge deal for a few reasons:

  • Fewer clicks means fewer abandoned carts. Buying right inside the AI conversation cuts down on the friction that makes people bail at checkout.
  • AI is the new storefront. For a lot of shoppers, the first “page” they see won’t be your homepage or your category page. It’ll be an AI’s answer that either includes your product or doesn’t.
  • Shoppers are going to be more specific. People talk to AI the way they’d talk to a knowledgeable store clerk. They’ll ask for exact models, sizes, compatibility and brands, and they’ll expect an exact match.  Let me repeat that because it’s incredibly important… people talk to AI the way they’d talk to a knowledgeable store clerk.

AI Can Only Sell What Your Data Describes

An AI shopping assistant can’t look at your product photo and “just know” what it is. It can’t call you up to ask which filter fits that model WRS325SDHZ fridge. It works off the data you’ve given it, and for Google, that’s mostly your Merchant Center feed, which is built from your product catalog.

When a person browses your site, they can fill in the blanks. They’ll read between the lines of a weak product description, squint at a photo you should have retaken years ago, or figure out the size from a product review. An AI agent won’t do that. If it isn’t confident your product matches what the shopper asked for, it’ll recommend someone else’s product that it is confident about.

So the stores that get picked aren’t necessarily the biggest or the cheapest. They’re the ones with complete, accurate, consistent product data. Messy data used to cost you a little SEO juice and a few disapproved items in Merchant Center. With AI driven checkout, it can mean you don’t show up at all.

Why UPCs, GTINs and MPNs Matter More than Ever

If there’s one piece of product data you shouldn’t skip, it’s your product identifiers. A UPC or GTIN is basically a product’s fingerprint. It’s the number that tells Google, “this exact item, in this exact size and color, from this exact manufacturer.”

So what’s the difference between the two? GTIN (Global Trade Item Number) is the umbrella term for the whole family of product barcodes, and a UPC is just one type of GTIN: the 12 digit version you’ll find on most products sold in the US and Canada. In other words, every UPC is a GTIN, but not every GTIN is a UPC, since the family also includes 13 digit EANs used in Europe and 14 digit codes for cases and multipacks.

A GTIN matters a lot when an AI is trying to match a shopper’s request to a real product. Product names vary from store to store. One shop calls it a “Blue Widget 12 oz,” another calls it “Widget, Blue, 12oz Bottle.” A GTIN cuts through all of that. It lets Google know your listing is the same product the shopper asked about, and it lets your offer compete directly with everyone else selling that item.

Google’s relied on these identifiers for years. Products that have a manufacturer assigned GTIN are expected to have it included in your feed.  Listings with missing or bad identifiers are more likely to get limited or disapproved in Shopping. As more buying shifts into AI Mode and Gemini, that same data becomes the thing that gets you matched (or not).

Some common problems we see all the time:

  • Blank UPC or GTIN fields, especially on products added in a hurry or imported from a supplier spreadsheet.
  • Junk in the field, like spaces, dashes, letters or placeholder text such as “N/A” or “none.”
  • Missing MPNs and brands, which Google uses to identify products when there’s no GTIN.
  • The same barcode on multiple variants, so every size or color looks like the same item.

There are Other Important Fields Too!

When an AI is deciding whether your product is the right answer, it’s looking at everything you’ve given it, along with the GTIN.  The “everything” should also include:

  • Brand or vendor. Shoppers ask for brands by name. If the field’s empty, you’re invisible to those searches.
  • Descriptions. A blank or one line description gives the AI nothing to work with when a shopper asks about features, materials or compatibility.
  • Images and alt text. Missing images hurt everywhere, and alt text is one more place to clearly describe what the product is.
  • Weight and dimensions. These drive shipping costs and delivery estimates, which matter a lot when checkout happens right inside the AI conversation.
  • Categories, collections and product types. These help both search engines and AI understand what kind of product you’re selling.
  • SKUs. Empty or duplicate SKUs cause headaches in feeds, inventory syncing and order processing.
  • Pricing sanity. A sale price that’s higher than the regular price (or a price above the compare at price on Shopify) looks broken to shoppers and machines alike.
  • SEO fields. Page titles and meta descriptions still help your products get found in regular search, which isn’t going anywhere.

None of this advice is new. What is new is the cost of getting it wrong!  When a shopper never visits your site, your data has to do all the selling, because remember, people talk to AI the way they’d talk to a knowledgeable store clerk.

Find the Gaps Fast with Missing Data Finder

Here’s the problem. If you’ve got a few hundred products, you might be able to eyeball them. If you’ve got a few thousand (or tens of thousands), there’s no way you’re clicking through every one to see what’s missing.

That’s exactly why we built Missing Data Finder as part of Your Store Tools. It scans your catalog, checks each product against the rules you pick, and hands you a report of everything that needs attention. It’s available for both BigCommerce and Shopify.

Here’s how it works:

  1. Pick your products. Scan your whole catalog, or narrow it down by SKU, category or collection, brand or vendor, stock status, visibility and more.
  2. Pick your checks. Choose which missing or invalid data you want to look for.
  3. Grab your report. The scan runs in the background, so you can close the page. When it’s done, you’ll get an optional email and a CSV or Excel file to download.

The report lists one row per issue, so if a product’s missing an image, a brand and a UPC, you’ll see three rows for it. That makes it easy to sort, filter and hand off to whoever’s fixing your catalog.

Screenshot of the missing data finder field selector

Missing Data Finder for BigCommerce

The BigCommerce version checks for empty UPC, GTIN and MPN fields, and it’ll flag UPC and GTIN values that contain anything other than numbers. It also looks for missing SKUs, names, descriptions, brands, categories, images and image alt text, plus weight and dimensions on physical products. On top of that, it catches sale prices higher than the regular price, default tax codes and classes, and empty page titles, meta descriptions and search keywords.

Missing Data Finder for Shopify

The Shopify version flags barcodes that are empty or invalid, empty and duplicate SKUs, and products missing descriptions, collections, product types, vendors or tags. It also finds products without images, images without alt text, variants without their own image, and prices higher than the compare at price.

One thing to know: Missing Data Finder is a report only tool. It won’t change anything in your store on its own, so you’re always in control of what gets fixed and how.

Don’t Wait to Make Changes!

UCP is still rolling out, and that’s actually good news for you. It means there’s time to get your catalog in shape before AI checkout becomes the norm. The stores that clean up their data now are going to be the ones AI assistants trust and recommend later.

Start with a scan. Run Missing Data Finder against your catalog, see where the gaps are, and knock out the biggest ones first. UPCs, GTINs, brands and descriptions are a great place to start.

And if your report comes back longer than you expected (it happens more than you’d think), we can help. The team at Your Store Wizards fixes product data, cleans up Google feeds and gets BigCommerce and Shopify stores ready for whatever’s next. Get in touch and let’s talk about your catalog.

Leave a Reply

Press ESC key to close search