Catalog ads

What catalog ads are

Let's take a shoe workshop in Mexico as an example. They make leather shoes and sell them online to Mexico and the United States. The shoe workshop has about 80 styles in the shop. Building 80 ads by hand, then updating prices by hand, does not survive a season. The catalog is the workaround: one feed, many ads.

One feed can power ads on several platforms at once: Meta catalog sets, Google Shopping and Performance Max (PMax: Google's automated campaign that mixes Search, Shopping, YouTube, Display, and more), TikTok catalog, Pinterest. The ad unit differs. The dependency is identical. Bad feed, bad ads.

What the feed has to get right

The feed is a file. Each product needs an ID, a human title, a price, an image, availability, and whatever extra labels you will later split on (margin, bestseller, season). Warehouse strings like OXF-TAN-42-W make terrible headlines. A tan leather oxford priced at $189 is what a person scanning a thumbnail can read.

Let's take a shoe workshop in Mexico as an example. They make leather shoes and sell them online to Mexico and the United States. The shoe workshop's first export used internal names. Google Shopping showed OXF-TAN-42-W next to the photo. Click-through rate (CTR: clicks divided by impressions) sat at 0.4%. The workshop rewrote titles for humans, kept the same product IDs, and the same products climbed to about 1.2% click-through rate (CTR) on Shopping. The shoes had not changed. The feed copy had.

• Titles written for a human scanning a thumbnail, not a warehouse code string.
• Prices and stock refreshed often enough that people do not click into a sold-out page.
• One clean image per product, cropped so the item survives at small sizes.
• Product types and custom labels you can actually segment on later.

Catalog ads across platforms

Same feed, different jobs. On Meta, catalog sets power DPA (dynamic product ads: personalized retargeting from what someone viewed) and broader prospecting. On Google, the feed drives Shopping and Performance Max (PMax) inventory. Splitting the catalog into sets by margin or bestseller status is usually more effective than another round of bid tweaks.

Let's take a shoe workshop in Mexico as an example. They make leather shoes and sell them online to Mexico and the United States. The shoe workshop spent $13,000 in a month across the feed: $4,000 Google Shopping and Performance Max (PMax), $3,000 Meta catalog prospecting, $6,000 Meta dynamic product ad (DPA) retargeting. Twelve styles took about 70% of that $13,000 spend ($9,100). The other 68 styles shared $3,900 and converted worse. The workshop split the catalog into a bestsellers set and a rest-of-range set so the twelve bestselling styles could have their own budget instead of starving everything else, and so the long tail was not judged on the bestsellers' return.

Common catalog ad mistakes

Uploading the ecommerce export untouched is the default. Let's take a shoe workshop in Mexico as an example. They make leather shoes and sell them online to Mexico and the United States. The shoe workshop also ignored Google's feed error report until 30 products were disapproved for missing identifiers or unusable images. Delivery dropped and it looked like a bid problem. It was a file problem.

• Uploading the ecommerce export untouched and calling it a creative strategy.
• Letting the whole catalog compete, so a handful of products absorb the budget.
• Ignoring feed error reports until delivery quietly drops.