Google Shopping disapprovals: fix attributes fast

You open Merchant Center, see a red count next to Products, and find that 78 of your 412 items are disapproved. Nobody told you. Shopping spend carried on, impressions quietly moved to the items that still qualified, and the ones you actually wanted to sell stopped appearing. The question is not whether to rebuild the feed. It is which attribute Google is missing, on which items, and how quickly you can put it back.

Direct answer: Disapprovals are nearly always an attribute problem, not a punishment. Merchant Center names the offending field in the issue string, so read it before you change anything: missing value gtin, missing value brand, mismatched value price and promotional overlay on image are four separate jobs with four separate fixes. Fix what Shopify can express at source in the bulk editor, patch what it cannot with a supplemental data source keyed on id, then request review. You do not need to rebuild the primary feed to clear most of them.

Read the disapproval before you touch the feed

The instinct when a feed breaks is to reach for a feed tool and start remapping. Resist it for an hour. In Merchant Center, open Products, then the Needs attention tab. Each row is an issue type, not a product, and clicking it gives you the affected item IDs and the exact attribute in square brackets. That bracket is the whole job. Missing value [gtin] is a data entry task. Mismatched value [price] is a storefront or currency task. Promotional overlay on image is an image production task. If you remap the feed without reading the bracket, you will spend a morning and clear none of them.

Then separate item level from account level. Item level issues affect a list of products and leave the rest of the account running. Account level policy issues, misrepresentation being the common one, take everything down at once and need a different flow: contact details, returns and refunds policy, and pricing that matches the landing page, followed by an account level review request. Do not confuse a promotional text warning on nine titles with a suspension.

The five reasons that account for most Shopify rejections

Shopify feeds fail in a predictable pattern because the Google and YouTube channel maps store fields to Google attributes, and store fields were never designed for Google.

Titles deserve one extra note. Google reads the title as sent, and shoppers see roughly the first 70 characters in the unit. A title stuffed with promotional claims fails policy, and a title that opens with your brand name and closes with the colour and size wastes the visible half. Run a sample through the title checker before you decide whether the fix is a policy edit or a merchandising rewrite.

A worked triage: 412 SKUs, 78 disapproved

Numbers make the priority obvious. Take an apparel store spending £150 a day on Shopping, with 412 items in the feed and 78 disapproved. Export the item IDs from Needs attention, then join them to the last 30 days of Shopping revenue in Google Ads by item ID. The picture usually looks like this.

Those 78 items are 19 per cent of the catalogue and £3,720 of 30 day revenue, which is £124 a day. At a 58 per cent gross margin that is about £72 of gross profit walking out of the door daily. Google's review after you request one takes a few working days, so the realistic cost of noticing on Thursday instead of Monday is roughly £500 of revenue. Notice the shape of the list too: 54 items carry half the money and take 40 minutes, while 8 image fixes take 90 minutes for a sixth of it. Do the gtin pass first, before lunch, and schedule the image work.

Fix at source, patch with a supplemental data source

Two routes, and they are not rivals. Source fixes are the ones Shopify can express: Barcode fills gtin, Vendor fills brand, variant options named Colour and Size fill colour and size, and the Google and YouTube channel picks all of them up on the next sync. Do these in Products, select all, Edit products, then add the Barcode and Vendor columns to the bulk editor. One pass, one afternoon, and they stay fixed for every channel that reads your catalogue, not just Google.

Everything else goes in a supplemental data source. In Merchant Center open Data sources, choose Add supplemental data source, and connect a Google Sheet with an id column plus only the columns you are correcting. Identifier exists set to no for handmade or own brand lines. A Shopping specific title that drops the promotional line while your storefront title stays as your SEO team wrote it. Age group and gender for apparel. Custom labels if you want to segment the recovered items into their own ad group while you watch them.

The rule that keeps this safe is that a supplemental source matches existing item IDs and corrects columns. It cannot invent products and it does not replace the primary feed the sales channel sends. That distinction is the difference between a reversible patch and a feed rebuild you cannot undo at 6pm on a Friday. The mechanics, including how to keep the sheet from rotting, are in supplemental feeds, done safely.

A disapproved SKU is not a warning. It is a product you have withdrawn from sale in the largest shop window you rent, without telling anyone.

What to do while you wait for review

Once the corrected values show on the product page inside Merchant Center, use Request review against the issue. Do not request review before the new data has actually been fetched, because a failed review makes the next one slower and you will have burned the request on the old file. While you wait, treat the campaign as short a few of its best items. If 54 gtin failures were your top sellers, your Performance Max or Standard Shopping campaign has been quietly reallocating budget to weaker items for however many days this ran, and the ROAS you are looking at is not the ROAS of the assortment you meant to sell.

Then work out how you missed it. In most stores nobody has a job that includes opening Merchant Center. The Shopify side does surface some of this: Settings, Apps and sales channels, Google and YouTube, then the Overview shows a product status count. It is coarser than the Merchant Center issues table, but a weekly glance catches the collapse rather than the drift. The cost of not looking is the same cost described in catalogue lag is a media tax: media keeps spending against a catalogue that stopped being true.

Do it yourself first

Set aside one afternoon and follow this order. First, in Merchant Center, open Products, then Needs attention, click each issue row and download the affected item IDs. Second, pull the last 30 days of Shopping performance by item ID from Google Ads and join the two lists in a sheet so every issue has a revenue figure next to it. Sort descending by revenue.

Third, take the source fixes. In Shopify go to Products, select all, choose Edit products, and add the Barcode and Vendor columns. Fill them for everything you can. For items with no manufacturer identifier, mark them on your sheet for the identifier exists patch. Fourth, in the Google and YouTube channel open the product feed settings and confirm colour, size, age group and gender are mapped from options or metafields. If age group and gender have no home in your product data, do not invent a metafield structure this afternoon; put them in the patch.

Fifth, build the supplemental sheet. One tab, an id column that matches the item IDs exactly, and only the columns you are correcting. Add it in Merchant Center under Data sources, Add supplemental data source, and link it to the primary feed. Sixth, wait for the fetch, confirm the new values appear on a sample product inside Merchant Center, then use Request review. Seventh, put a recurring 10 minute check in the calendar on Monday mornings for the Needs attention count, and note the number. A number that moves from 3 to 40 is the signal, not the absolute value.

Where Ralph fits

Ralph scans product attributes and Shopping readiness, and bulk attribute fixes are gated with a cost ceiling and go out as supplemental feeds only, never as a silent overwrite of your primary feed. Honestly: he will not argue an account level misrepresentation suspension for you, he will not re-shoot an image that has a sale badge burned into it, and no write happens until you press approve. Ralph is not open for self serve use yet, so the waitlist is the only door.

What this is not

This is not the same problem as items dropping out of Shopping because they sold out. A disapproval is a data or policy failure that Google reports in the issues table. An out of stock item is correctly excluded and reappears when stock returns, which is a different decision covered in Google Shopping when a size sells out. It is also not a feed management project. Full feed platforms rebuild and transform your primary source, which is a reasonable purchase at large catalogue sizes, but reaching for one to fix 54 empty barcode fields is buying a warehouse to store a shoebox. Fix the barcodes, patch the rest, and see what is left before you sign anything.

Fix disapproved products in Google Merchant Center

1. Export the item issues. In Merchant Center open Products, then Needs attention. Click each issue to list affected item IDs and download the list.
2. Rank by money. Join the item IDs to the last 30 days of Shopping revenue in Google Ads so you fix the earners before the long tail.
3. Fill identifiers in Shopify. Go to Products, select all, then Edit products, and add the Barcode and Vendor columns so gtin and brand are populated.
4. Map apparel attributes. In the Google and YouTube channel feed settings, map colour, size, age group and gender from variant options or metafields.
5. Patch the remainder. In Merchant Center open Data sources, add a supplemental data source keyed on id, and include only the columns you are correcting.
6. Request review. Once the corrected values appear on the product in Merchant Center, use Request review on the affected issues.
7. Verify in three days. Reopen Needs attention and confirm the items now read Eligible rather than Not eligible before closing the job.

If you want stock, ads and the morning list in one place: that is what we built Ralph for. A growth operator for Shopify, with review before anything ships. Not magic. Fewer tabs. See Ralph · Docs