What is Product Feed?
Also known as: Product data feed, Shopping feed, Catalog feed
What is a product feed?
A product feed is a structured data file that lists every product a retailer sells, formatted to match an ad platform's required schema. Per Google's product data specification, each row carries attributes like id, title, description, price, availability, and image_link. The feed is the source of truth for Google Shopping ads and catalog ads.
Think of it as the bridge between a store and an ad network. Shopify holds the inventory. Google, Meta, Pinterest, and TikTok need that data in their own format to render product cards. The feed translates one to the other.
[UNIQUE INSIGHT] The feed is the creative. On Shopping and dynamic product ads, the advertiser writes no headlines and picks no keywords. The product title, image, and price are the entire ad. Optimizing the feed is optimizing every impression.
Required vs recommended attributes
Every channel has a non-negotiable core. Per Google Merchant Center's product data spec, nine attributes are required for most retail products in 2026, with brand and GTIN required for any branded item.
| Attribute | Required | What it does |
|---|---|---|
id | Yes | Unique SKU identifier across the feed |
title | Yes | Product name shown on the ad card, 150 char max |
price | Yes | Must match the landing page price exactly |
availability | Yes | in stock, out of stock, preorder, backorder |
image_link | Yes | Direct URL to a clean product image, 800px+ |
gtin | Yes (branded) | UPC, EAN, or ISBN for the manufacturer code |
brand | Yes (branded) | Manufacturer brand, exempt for custom or generic |
condition | Yes | new, refurbished, or used |
google_product_category | Recommended | Google's taxonomy for matching to queries |
Recommended attributes change which queries the product matches on. Color, size, gender, age_group, material, and pattern all feed extra signal. Custom labels (custom_label_0 through 4) let advertisers segment by margin tier, season, or best-seller status.
Where do product feeds go?
Product feeds power every major shopping ad surface. The big four channels each have their own catalog manager and submission method, but the underlying data model is similar.
| Platform | Catalog hub | Feed format | Refresh cadence |
|---|---|---|---|
| Merchant Center | XML, TXT, Google Sheets, Content API | Hourly to daily | |
| Meta (Facebook, Instagram) | Commerce Manager catalog | CSV, TSV, XML, Pixel-based, API | Hourly to daily |
| Pinterest catalog | CSV, TSV, XML | Daily | |
| TikTok | TikTok Catalog Manager | CSV, XML, API | Hourly to daily |
Most ecommerce platforms automate the export. Shopify's product feed guide ships a Google channel app that builds and syncs the feed. The same pattern applies to the Shopify ads integration and the WooCommerce ads integration. Headless storefronts usually pipe through a feed management tool.
Feed quality and disapproval reasons
Feed errors stop ads from serving. Per Google Merchant Center diagnostics data referenced in Search Engine Land's 2024 reporting, roughly 15 to 25 percent of products in a typical retail feed carry at least one warning or disapproval at any time.
The five most common reasons:
- Price mismatch. Feed says $49.99, landing page says $54.99. Google catches the gap on a crawl and disapproves the SKU.
- Missing GTIN or brand. Required on any branded item. Use
identifier_exists: falsefor genuine custom or handmade goods only. - Image policy violations. Promotional text, watermarks, logos, or borders burned into the image. The primary must show the product alone.
- Inaccurate shipping or returns. Feed claims free returns, site policy says otherwise. Triggers a misrepresentation strike.
- Landing page issues. Broken link, redirect loop, or a product that's actually out of stock when the feed says in stock.
[PERSONAL EXPERIENCE] In our experience auditing retail accounts, fixing price mismatches and adding missing GTINs alone rescues 40 to 60 percent of disapproved SKUs in the first review cycle. The fix lives in the feed, not the campaign.
Feed optimization tactics
Feed quality drives impression share more than bid strategy does. Per Search Engine Journal, retailers who run quarterly feed audits see 20 to 40 percent more impressions on the same budget.
Title structure
Front-load the highest-value terms. The first 70 characters carry most of the matching weight. Working patterns by category:
- Apparel. Brand + Product Type + Gender + Color + Size + Material
- Electronics. Brand + Model + Key Spec + Color
- Home goods. Brand + Product Type + Material + Size + Color
Stuffing the title with marketing fluff ("Premium Quality," "Best Seller") wastes the prime real estate.
GTIN coverage
Every branded SKU should carry a valid GTIN. Products with GTIN see higher impression share because Google can match them to specific product knowledge graph entries. For private-label products without a manufacturer code, set identifier_exists: false and provide brand plus MPN.
Custom labels for campaign segmentation
Custom labels split the catalog into bid-worthy groups. Common patterns:
custom_label_0: margin tier (high, mid, low)custom_label_1: season (spring, summer, fall, winter)custom_label_2: best-seller flagcustom_label_3: clearance or full-pricecustom_label_4: new arrival window
Inside Google Ads, build separate campaigns or product groups by label so high-margin SKUs get higher Target ROAS than clearance. The same labels feed Performance Max asset groups when scaling beyond Standard Shopping.
Real-world example with numbers
A home goods retailer with 4,300 SKUs runs a Performance Max campaign on a $500 daily budget. Feed audit week one shows 28 percent of products disapproved or warning.
[ORIGINAL DATA] After a structured feed cleanup over 14 days:
- Disapprovals dropped from 1,204 SKUs to 89.
- Active impression-eligible SKUs jumped from 3,096 to 4,211.
- Daily impressions rose 64 percent on the same budget.
- ROAS climbed from 2.1 to 3.4 over four weeks.
The fixes were unglamorous. Adding GTINs to 720 branded items. Rewriting 1,800 titles to lead with brand and product type. Replacing 340 lifestyle images with white-background shots. No bid changes. No new creatives. Just feed hygiene.
Product feeds in 2026
Three shifts decide whether a feed wins this year.
AI-generated attributes are mainstream. Tools now auto-write titles, descriptions, and category mappings from a single product image plus a name. Quality varies. Human review on the top 20 percent of revenue SKUs still beats fully automated output, but the long tail benefits from AI enrichment.
Multi-channel feeds are table stakes. Running only Google Shopping leaves Meta Advantage+ catalog ads, Pinterest shopping pins, and TikTok shop ads on the table. A feed management layer that maps one source to four channel destinations is now standard for retailers over $1M in annual ad spend.
Real-time signal beats nightly batch. Stock-out clicks waste budget and erode quality. Retailers moving from once-daily feed pushes to Content API or supplemental near-real-time updates report 10 to 15 percent fewer wasted clicks per Merchant Center benchmarks. Speed of signal is the new optimization frontier.
The feed is the foundation. Bid strategy is the easy part.
Related terms
Frequently asked questions
What is the difference between a product feed and a product catalog?
A catalog is the master list of products in a system like Shopify or WooCommerce. A product feed is the formatted export of that catalog sent to an ad platform. The catalog is the source. The feed is the channel-specific delivery, mapped to that platform's required attributes and refreshed on a schedule.
How often should a product feed update?
Daily at minimum. Twice a day for retailers with fast inventory turnover, hourly for flash sales or marketplace sellers. Google Merchant Center accepts scheduled fetches up to every hour and the Content API for near real-time pushes. Stale prices and stock status are the top two disapproval triggers per Google's product data spec.
Do I need a different feed for each ad platform?
Yes, in practice. The base attributes overlap (id, title, image, price), but Google requires GTIN and Google product category, Meta requires availability and condition in its own format, and TikTok wants its taxonomy. Most retailers use a feed management tool (DataFeedWatch, Feedonomics, GoDataFeed) to map one source to many channel feeds.
Why are products getting disapproved in Merchant Center?
Most disapprovals come from four issues. Mismatched price between feed and landing page. Missing GTIN or brand on branded products. Promotional overlays burned into the image. Inaccurate shipping or returns data. Fix the feed first, then request review. Repeat policy violations lead to account suspension.
Can I run Shopping ads without a product feed?
No. Every Shopping campaign on Google, every Advantage+ catalog ad on Meta, and every dynamic product ad across networks pulls from a feed. Performance Max requires a Merchant Center feed for retail goals. Even AI ad platforms that auto-generate creatives still read product data from a feed to populate titles, prices, and images.