Back to blog

Lecture 17: Shopping Campaigns and Product Feed Optimization

SEM Course

Lecture 17: Shopping Campaigns and Product Feed Optimization

By Maya | Search Engine Marketing Strategist

Lecture 17 of the Complete SEM Mastery course: learn how Google Shopping campaigns work, how to build and optimize a Merchant Center product feed, write high-converting titles, fix disapprovals, and improve impression share through feed quality.

Complete SEM Mastery, Lecture 17 of 30

A 30-lecture course covering paid search strategy from keyword research to advanced automation, built for marketers who want to run profitable Google Ads accounts with confidence.

Back to Lecture 16 | Go to Lecture 18 | Run a Free SEO Audit

Short answer: Shopping campaigns do not run on keywords you bid on directly. Google matches your product feed data (title, description, category, attributes) against search queries, meaning your ranking and click-through rate depend almost entirely on how well-structured, complete, and keyword-rich your Merchant Center feed is. Master the feed, and you control which queries trigger your products, how your listings look on the results page, and how efficiently your budget is spent.

What You'll Learn in This Lecture

  • How Shopping ads fundamentally differ from text search ads in targeting and ranking
  • Step-by-step Google Merchant Center account setup and verification
  • The core product feed structure and every required attribute
  • How to write product titles that maximize both relevance and clicks
  • Description, image, and identifier (GTIN/MPN) requirements that affect approval
  • How to use feed rules and supplemental feeds to optimize at scale without touching your source feed
  • How campaign priority settings and shared negative keyword lists control traffic
  • The difference between Standard Shopping, Smart Shopping, and Performance Max
  • How to read Merchant Center diagnostics and fix the most common disapprovals
  • Techniques to raise feed quality score, impression share, and CTR
  • Bid strategy considerations specific to product-based auctions
  • How to structure product groups and subdivisions for granular bidding
  • A practical monthly feed audit checklist

How Shopping Ads Differ From Text Search Ads

In a standard Search campaign, you select keywords, write ad copy, and Google matches your keywords to a user's query. In Shopping campaigns, there are no keywords to bid on at all. Instead, you upload a structured product feed to Google Merchant Center, and Google's algorithm parses your product titles, descriptions, product type, and attributes to decide which search queries your product should appear for. This is called query-to-product matching, and it happens automatically based on the text and data you provide.

This distinction changes your entire optimization approach. In text ads, you write persuasive copy and select match types. In Shopping, your \"copy\" is your product title and description, and your \"targeting\" is the completeness and accuracy of your feed attributes. A poorly titled product with missing attributes simply will not surface for many relevant searches, no matter how high you bid. Conversely, a well-optimized feed entry can outrank a competitor's identical product purely because Google's algorithm judges it more relevant and higher quality.

Another key difference is the auction itself. Shopping ads compete on a combination of bid, feed quality, historical CTR of the product, price competitiveness, and landing page experience. Google also displays multiple competing products side by side in a visual grid, so your product image, price, and any merchant promotions (free shipping badges, ratings, sale price strikethroughs) directly affect click-through rate in a way that text ads never do.

Example: Two stores sell the same running shoe. Store A titles the product \"Shoes\" with a blurry stock photo. Store B titles it \"Men's Nike Air Zoom Pegasus 40 Running Shoes - Black/White, Size 7-13\" with a clean lifestyle image and an accurate GTIN. Store B will show up for far more long-tail queries like \"nike pegasus 40 black size 10\" and will earn a materially higher CTR at the same bid, simply because the feed does the targeting work that keywords would do in a text campaign.

Google Merchant Center Setup Basics

Before any product can appear in Shopping ads, it must live in Google Merchant Center (GMC), the free platform where you manage your product data separately from Google Ads. Setup involves several sequential steps that are easy to get wrong.

First, create a Merchant Center account at merchants.google.com using a business email, ideally one tied to your company's Google Workspace domain rather than a personal Gmail address, since this affects trust signals during verification. Next, complete the business information section: legal business name, address, customer service contact, and return policy. Google increasingly requires an accurate, working returns and refund policy page linked from your site before it will fully serve ads, especially for accounts targeting the US and EU.

Then verify and claim your website URL. Google offers several verification methods: uploading an HTML file to your server, adding a meta tag to your homepage header, using Google Tag Manager, or verifying through Google Analytics or Google Search Console if already connected. Claiming confirms you have authority over the domain; verification confirms you own it. Both are required before products can go live.

Next, set up tax and shipping settings at the account level. This is mandatory in most countries before products are approved. You can configure flat-rate shipping, carrier-calculated rates, or free shipping thresholds, and Google will display this shipping cost directly on the Shopping listing, so accuracy matters for both compliance and conversion rate. Incorrect shipping settings are one of the top causes of price/shipping mismatch disapprovals.

Finally, link Merchant Center to your Google Ads account under Settings > Linked accounts. This link is what allows you to build Shopping campaigns pulling from your feed. Without this link, your feed exists but no ads can run against it.

Product Feed Structure and Required Attributes

The product feed is a structured file, submitted as a spreadsheet (Google Sheets), a scheduled file fetch (CSV/XML/TSV hosted on your server), or via the Content API for larger catalogs with frequent updates. Each row represents one product variant, and each column is an attribute Google reads to understand and categorize that product.

The required attributes for most product categories are: id (a unique, permanent identifier for each product/variant), title, description, link (the product landing page URL), image_link, availability (in stock, out of stock, preorder), price, brand, condition (new, refurbished, used), and identifiers like gtin and/or mpn. Depending on category, Google also requires google_product_category and product_type to correctly classify items, plus category-specific attributes such as size, color, gender, age_group, and material for apparel.

Recommended (non-required but heavily impactful) attributes include sale_price, sale_price_effective_date, additional_image_link (up to 10 extra images), shipping_weight, custom_label_0 through custom_label_4 (used for segmenting campaigns by margin, bestseller status, or season), and promotion_id for tying in Merchant Promotions. Custom labels deserve special attention: they don't affect matching at all, but they are the backbone of smart campaign structure, letting you bid differently on high-margin versus low-margin SKUs, or bestsellers versus long-tail inventory, without touching titles or descriptions.

Feed data should be refreshed at least daily, and ideally through an automated scheduled fetch rather than manual upload, since stale price or stock data leads directly to disapprovals and poor user experience (a shopper clicking through to an out-of-stock item is a wasted click and a damaged Google trust signal).

Writing Optimized Product Titles for the Feed

The title attribute is the single most influential field for both matching and click-through rate. Google truncates titles at roughly 150 characters in the feed, but only the first ~70 characters reliably display on most devices, so front-load the most important, searchable terms.

A reliable title formula for most e-commerce verticals is: Brand + Product Type + Key Attributes (color, size, model) + Distinguishing Feature. For apparel: \"Brand + Gender + Product Type + Material + Color + Size.\" For electronics: \"Brand + Model Number + Product Type + Key Spec.\" Avoid generic titles like \"Blue Dress\" and avoid keyword-stuffing irrelevant terms, since Google's algorithm and its policy team both penalize titles that look manipulative rather than descriptive.

Do not waste title characters on promotional language like \"Best Price\" or \"Free Shipping!\" — those belong in sale_price and shipping attributes, where Google can display them as actual badges, which perform far better than text claims buried in a title. Also avoid ALL CAPS and excessive punctuation, both of which can trigger a policy review or simply look untrustworthy in the ad grid next to competitors using clean titles.

Example: Instead of \"Amazing Comfortable Yoga Pants for Women Best Quality,\" use \"Women's High-Waist Yoga Leggings - Black, Squat-Proof, Size XS-XXL.\" The second version surfaces for searches like \"high waist yoga leggings black\" and \"squat proof leggings XL\" that the first title would never match, because Google is matching on concrete, searchable nouns and modifiers, not adjectives.

Product Descriptions, Images, and GTIN/MPN Requirements

The description field (up to 5,000 characters, though only the first several hundred meaningfully influence matching) should read as a natural, informative product description including material, use case, size details, and care instructions — written for a human shopper, not stuffed with keywords. Google does use description text for matching secondary and long-tail queries the title doesn't cover, so it is worth writing uniquely per product rather than copy-pasting a generic paragraph across your entire catalog.

Image requirements are strict: no watermarks, no promotional text or banners overlaid on the image, no placeholder or \"coming soon\" graphics, and a minimum resolution (100x100px for most categories, 250x250px recommended, with apparel requiring at least 250x250px and ideally 800x800px or larger for zoom quality). Lifestyle and in-context images generally out-convert plain white-background product shots for apparel and home goods, while electronics and hard goods often perform better with clean, well-lit studio images that clearly show the product's form factor.

GTIN (Global Trade Item Number — UPC, EAN, ISBN, or JAN depending on region) and MPN (Manufacturer Part Number) are how Google validates that your product is a real, recognized item and matches it against its own product catalog. For most branded products, submitting an accurate GTIN is now effectively mandatory — Google has increasingly restricted or limited impressions for products missing valid GTINs when one exists for that item. If you sell a private-label or custom-manufactured product with no manufacturer GTIN, set identifier_exists to \"false\" so Google doesn't flag the missing identifier as an error; if a GTIN does exist for the item, submitting the wrong one (or none at all) is one of the most common causes of suspended or disapproved products.

Feed Rules and Supplemental Feeds for Bulk Optimization

For catalogs with hundreds or thousands of SKUs, manually editing every title or description in your source system is impractical. Merchant Center offers two tools for optimizing at scale without touching your primary product data: feed rules and supplemental feeds.

Feed rules let you transform data on the fly as it's ingested into Merchant Center — for example, automatically prepending the brand name to every title if it's missing, appending size and color to titles that lack them, or stripping HTML tags from descriptions pulled from a CMS. Rules are built with a simple condition-and-action interface (or via the Content API for advanced logic) and apply across your entire catalog or filtered subsets, saving enormous manual effort.

Supplemental feeds are separate files that merge with your primary feed to add or override specific attributes without modifying your main product database. This is especially useful when your e-commerce platform's native feed export lacks fields Google needs (like custom_label or additional_image_link) or when you want marketing-controlled overrides — for instance, a supplemental feed maintained by the PPC team that adjusts custom_label_0 values for margin-based bidding, while the primary feed stays owned by the web/dev team and reflects live inventory and pricing untouched.

Example: An online furniture retailer's primary feed, generated by their Shopify export, has generic titles like \"Sofa - Grey.\" Using a supplemental feed keyed to product ID, the marketing team adds enriched titles (\"Modern 3-Seater Sofa - Grey Linen, Sleeper Convertible\") and custom labels flagging bestsellers, without ever asking the dev team to change the underlying product database.

Campaign Priority and Negative Keywords in Shopping

Even though Shopping campaigns don't use keyword bidding, you can and should apply negative keywords to prevent your products from matching irrelevant queries — for example, adding \"parts,\" \"repair,\" or \"free\" as negatives if you sell finished products, not components, spare parts, or giveaways. Negatives can be applied at the campaign or ad group level and are essential for controlling wasted spend, since query matching is algorithmic and occasionally overly broad.

Campaign priority is a Shopping-specific setting (Low, Medium, High) used when you run multiple Shopping campaigns targeting overlapping products. When the same product exists in more than one campaign, Google uses priority to decide which campaign competes in the auction — the highest-priority campaign with budget remaining wins, regardless of bid amount. This is commonly used to structure campaigns by intent: a low-priority \"catch-all\" campaign with broad coverage and lower bids, and a high-priority campaign for bestsellers or high-margin products with more aggressive bids and a dedicated negative keyword list to exclude everything else, ensuring your best products always get budget priority even when overall campaign budgets are constrained.

Standard Shopping vs Smart Shopping/Performance Max Migration

Standard Shopping campaigns give you full manual control: you set bids at the product group level, choose bidding strategy, control placements (Shopping tab, search results, images, YouTube if opted in), and see full search term reports. Smart Shopping campaigns, Google's now-deprecated automated hybrid format, combined Shopping and remarketing/display inventory under automated bidding with minimal reporting visibility — Google fully sunset Smart Shopping and required migration into Performance Max by 2022.

Today, most advertisers run either Standard Shopping (for full control, granular product group bidding, and detailed search term data) or Performance Max, which folds Shopping inventory into an automated, cross-network campaign type covering Search, Display, YouTube, Discover, Gmail, and Maps, driven by machine learning and asset groups rather than manual product group bids. Google has been steadily nudging advertisers toward Performance Max, and in many accounts it now auto-suggests or defaults new Shopping-oriented campaigns into Performance Max structure. The tradeoff is control versus reach: Standard Shopping gives you visibility into exactly which queries and products drive results, while Performance Max trades that transparency for broader automated reach and, in Google's own benchmarks, often higher conversion volume at similar CPA once the algorithm has enough conversion data to optimize against. We cover Performance Max in full depth in the next lecture, since it deserves dedicated treatment beyond the Shopping-specific scope here.

A sound strategy for many accounts is to run both in parallel with Standard Shopping set to exclude the products or search intents you want Performance Max to own, using priority and negatives to prevent internal competition, then compare performance over a full sales cycle before consolidating budget toward whichever structure delivers better feed-driven ROAS for your catalog.

Merchant Center Diagnostics and Common Disapprovals

The Diagnostics tab in Merchant Center is your primary health dashboard, flagging account-level issues, feed-level warnings, and item-level disapprovals with specific policy citations. Reviewing this weekly (not just when traffic drops) catches problems before they compound into major impression loss.

The most common disapproval categories are: missing or invalid identifiers (GTIN doesn't match the manufacturer's actual registered code, often from mistyping or using an internal SKU instead), price/shipping mismatch (the price or shipping cost shown on your landing page doesn't match the feed, usually caused by promotional pricing on-site not being reflected in the feed, or currency/tax handling differences), policy violations for images (watermarks, placeholder graphics, or promotional text baked into the image), misrepresentation (title or description overstating claims, like \"cures\" for supplements or unverified \"#1 rated\" claims), and landing page errors (broken links, redirect chains, out-of-stock pages, or pages requiring login before showing the product).

Account-level suspensions are more severe and typically stem from unverified/unclaimed URLs, missing business information, broken checkout or return policy pages, or repeated policy violations across many items signaling a systemic feed quality problem rather than isolated errors. When an account suspension occurs, Merchant Center provides an appeal path, but the fix must address the root cause across the whole catalog, not just the flagged sample items, or the appeal will fail.

Example: A supplement brand's feed showed \"price/shipping mismatch\" disapprovals across 40% of SKUs. Investigation revealed the site displayed prices with a member discount applied by default (via a stored browser cookie), while the feed submitted the full retail price — Google's crawler, without that cookie, saw a different price than the feed declared. The fix was submitting the actual first-visit price shoppers see, resolving the mismatch account-wide within 48 hours of the next feed refresh.

Optimizing Feed Quality to Improve Impression Share and CTR

Feed quality directly drives impression share, the percentage of eligible auctions where your product actually shows. Low-quality feeds — thin titles, missing attributes, weak images — get filtered out of many auctions entirely before bid even becomes a factor. Improving feed quality is therefore often the highest-leverage lever in a Shopping account, ahead of bid increases.

Start by running the Search terms report at the campaign level to identify high-volume queries your products should be matching but aren't; missing matches usually trace back to a title or product_type gap. Next, audit your google_product_category and product_type fields for accuracy — misclassified products (e.g., a phone case listed under \"Cases\" generically instead of \"Electronics > Accessories > Cases > Phone Cases\") get shown in the wrong auctions and suppressed in the right ones.

Use the Best Sellers and Price Competitiveness reports in Merchant Center to see how your pricing and assortment compare against the category benchmark; products priced significantly above competitors for an identical GTIN will lose impression share even with a well-optimized feed, since Google factors relative price competitiveness into the auction. Enable Merchant Center ratings and reviews feeds where available, since seller ratings and product review stars displayed directly in the ad meaningfully lift CTR at no additional bid cost.

Finally, segment performance by custom_label to isolate which product tiers (bestsellers, clearance, new arrivals, high-margin) are driving efficient spend, and reallocate budget and bids toward the tiers with the strongest ROAS, using product group subdivisions in Standard Shopping campaigns to bid up your winners and bid down or exclude chronic underperformers rather than applying one blanket bid across the whole catalog.

Treat the feed as a living asset that needs the same ongoing optimization discipline as ad copy or landing pages — a Shopping account with even a modest 500-SKU catalog can see meaningful impression share and conversion rate gains from a single focused month of title rewrites, image upgrades, and attribute cleanup, often outperforming a bid-only optimization approach by a wide margin.

Related Lessons Across SEO, AEO, GEO, SEM, and PPC

Use these connected lessons to move through organic search, answer engines, generative AI visibility, paid search, and PPC without losing the bigger strategy.