Product Feed Optimization for Google Merchant Center: The Complete Field-by-Field Guide

Product feed optimization is the practice of structuring and enriching your product data in Google Merchant Center so that Shopping ads, Performance Max, and unpaid listings match more searches, win more clicks, and convert at a higher return. It matters because for Shopping and Performance Max there are no keywords you bid on. Google reads your feed and decides which searches your products deserve. A clean, keyword-rich, accurate feed is the single biggest lever on impressions, click-through rate, and return on ad spend. A weak feed silently throttles good products and burns budget on the wrong queries.

This guide walks every field that moves the needle, in the order it matters, with formulas and a concrete before-and-after example.

Why is the product feed the campaign for Shopping and Performance Max?

In a Search campaign you write keywords and ads. In Shopping and Performance Max you write neither. There is no keyword list. Google's systems parse each product's title, description, category, attributes, and image, then infer which queries the product should show for. Your feed is your keyword list, your ad copy, and your quality signal rolled into one data file.

This changes the whole optimization model. You do not tune bids toward a query. You improve the data so Google's matching engine understands the product, then let automated bidding do its job. When a feed is thin, Google guesses, which means irrelevant impressions, low click-through rate, and a weaker quality signal that pushes cost per click up. When the feed is rich and precise, matching tightens, the right shoppers click, and return on ad spend climbs.

This is why feed work sits upstream of everything else in ecommerce PPC. You can have perfect bids and a healthy budget, but if the feed is wrong the ceiling is low.

How do you optimize the product title for Shopping and Performance Max?

The title is the highest-impact field in the entire feed. It carries the most weight in Google's matching and it is the text shoppers read before they click. You get 150 characters. Use them deliberately.

Front-load the most important keywords into the first 70 characters, because that is roughly what shows in most placements before truncation. Match how people actually search, not your internal product names.

A reliable formula, adapted by category:

• Apparel: Brand + Product Type + Attributes + Color + Size + Gender. Example: Nike Air Max 270 Running Shoes Black Mens Size 10
• Electronics: Brand + Model + Product Type + Key Spec + Color. Example: Samsung Galaxy S24 Ultra 256GB Smartphone Titanium Black
• Home goods: Brand + Product Type + Material + Size + Color. Example: IKEA Malm 6-Drawer Dresser Oak Veneer 160cm White
• Beauty: Brand + Product Line + Product Type + Size/Volume + Skin Type. Example: The Ordinary Niacinamide 10% Serum 30ml Oily Skin

Rules that hold across categories:

• Put the brand and the core product type first. Shoppers and Google both anchor on these.
• Include the attributes people search: color, size, material, capacity, model number, gender, quantity.
• Avoid promotional text. Words like "sale," "best price," "at no media cost," or "fast shipping" are disallowed in titles and waste characters that could carry keywords.
• No block capitals, no gimmick symbols, no keyword stuffing. Write it the way a careful shopper would type it.
• Keep it truthful and specific. A vague title matches vague, low-intent queries.

For a store on Shopify or Salla selling into the UAE, that also means writing titles for how local shoppers search, mixing brand names with the practical attribute (size, capacity, compatibility) that decides the purchase.

What is the difference between a wasteful title and a strong one?

A bad title costs money in two ways. It matches queries that do not fit, so you pay for clicks that never convert. And it fails to match the queries that would convert, so you never appear where the intent is highest. Both drag return on ad spend down at once.

Here is a concrete before-and-after for a single product.

Before: Cotton T-Shirt

After: Levi's Original Crewneck Cotton T-Shirt Men's Navy Blue Size L

The first title tells Google almost nothing. It competes in a vast, generic pool, matches broad low-intent searches, and gives shoppers no reason to click over a more descriptive rival. The second carries the brand, product type, material, gender, color, and size. It now matches high-intent searches such as "Levi's navy crewneck men's L," which are cheaper to win and far more likely to convert. Practitioners who make this kind of change across a catalog commonly report meaningful lifts in impressions and click-through rate, and a stronger return on ad spend, because spend shifts from generic misses to qualified matches. The exact numbers vary by catalog and category, but the direction is consistent.

How should you write the product description?

The description is a secondary matching field and a conversion tool. Google reads it to reinforce relevance, so the first 160 to 500 characters should carry the important details and keywords, since the top of the description is weighted most.

Best practice:

• Lead with the key features, materials, dimensions, use cases, and compatibility.
• Write in full sentences with natural keyword coverage. Do not stuff.
• Include attributes that did not fit the title: care instructions, what is in the box, warranty, sizing notes.
• Keep it accurate to the landing page. Mismatches between feed and page hurt trust and can trigger disapprovals.
• Skip promotional language and store policies. This field is about the product, not the offer.

The full description limit is 5,000 characters. You rarely need it all, but the richer and more accurate it is, the better Google understands edge-case queries.

How do google_product_category and product_type work together?

These two fields tell Google what the product is, and they play different roles.

google_product_category is Google's own fixed taxonomy. You pick the single most specific node that fits, for example "Apparel & Accessories > Clothing > Shirts & Tops." Being specific helps Google place the product correctly and can affect which formats and features it qualifies for. Some categories require extra attributes such as size, color, age group, and gender, so accurate categorization also unlocks fields you should be filling.

product_type is your own taxonomy, taken from how your store is organized, for example "Home > Mens > Shirts > Crewneck." It is not shown to shoppers, but it is powerful for campaign structure. You can subdivide Shopping and Performance Max by product_type to control which products sit in which asset groups or listing groups, and to isolate winners. Set both fields on every product. They cost nothing and they sharpen both matching and structure.

Which product identifiers matter: GTIN, brand, MPN, and identifierExists?

Unique product identifiers let Google know exactly which product you are selling and connect it to its wider knowledge of that item, which improves matching and eligibility.

• gtin: the global trade item number, usually the barcode (UPC, EAN, ISBN). Provide it whenever the product has one. Correct GTINs are one of the strongest relevance signals available and are effectively expected for branded, manufactured goods.
• brand: the recognized brand name. Required for most categories.
• mpn: the manufacturer part number. Use it when there is no GTIN, or alongside brand for products where GTINs do not apply.
• identifierExists: set this to "no" only for genuinely custom, handmade, or own-brand products that truly have no GTIN or MPN. Do not use it as a shortcut to skip valid identifiers. Suppressing real identifiers weakens matching and can limit where products show.

Get these right and Google can associate your product with real demand. Get them wrong or omit them, and you look like a generic listing competing on price alone.

What are the image rules for Merchant Center?

The main image is the product's storefront. Google enforces standards, and a strong image directly lifts click-through rate.

• Use a plain white or light, uncluttered background for the main image where the category allows.
• No promotional overlays, no logos on top of the product, no watermarks, no borders, no "sale" badges, no call-to-action text. These can cause disapproval.
• Show the actual product clearly, filling most of the frame without being cut off.
• Meet the minimum size. Non-apparel images should be at least 100 by 100 pixels and apparel at least 250 by 250 pixels, but that is the floor. Aim for at least 800 pixels on the longest side, and 1,000 to 1,500 pixels for a crisp result across placements.
• Match the image to the variant. A blue shirt listing should show the blue shirt.

Use additional_image_link to add more angles, lifestyle context, and detail shots. You can add several. These support the shopping experience and give Google more to work with, while the main image stays clean and compliant.

Why do price and availability accuracy matter so much?

Price and availability are the two fields that must always match your landing page exactly. Google crawls your site and compares. Any mismatch, even a small one, risks disapproval or a suspended account, and it erodes shopper trust when the price on the site differs from the ad.

• Keep price and availability in sync with the store in real time. Frequent price or stock changes call for a faster refresh, through Merchant Center's crawl settings, scheduled fetches, or the Content API.
• Use sale_price for the discounted amount and keep price as the regular amount. This is how Google shows a strike-through and a "sale" annotation, which lifts click-through rate honestly, without any banned overlay text on the image.
• Set sale_price_effective_date to bound the promotion window so the sale price applies only while it should.
• Mark out-of-stock items accurately. Advertising products you cannot sell wastes spend and frustrates buyers.

Reliable, near-real-time price and stock data is one reason feed health is tied to your ecommerce platform build. A clean data layer at the store level makes an accurate feed far easier to maintain.

How do custom labels segment campaigns for better control?

Custom labels are five optional fields, custom_label_0 through custom_label_4, that carry your own tags. Shoppers never see them. Their entire purpose is to let you subdivide Shopping and Performance Max so you can bid, budget, and report by the dimensions that matter to your business.

Common and effective uses:

• Margin: tag products as high, medium, or low margin, then push budget toward the profitable ones.
• Bestseller status: tag proven sellers so you can scale them separately from the long tail.
• Season: tag "summer," "ramadan," "back-to-school," so you can turn seasonal ranges up and down cleanly.
• Price band: tag "under-100," "100-500," "premium," so you can manage very different buyer behavior in separate groups.
• Performance tier: tag "new," "clearance," "hero," to control how much each tier gets.

With these labels in place, Performance Max and Shopping stop being one undifferentiated pool. You can concentrate spend where the return is, and starve or exclude the products that quietly waste it. Without labels, you are stuck managing the whole catalog as one blunt object.

What do product highlights, supplemental feeds, and extra attributes add?

Beyond the core fields, several enrichments add real value.

• product_highlight: short bullet points that summarize the best features. These can appear in the shopping experience and give Google more structured signal.
• Attribute completeness: fields such as color, size, material, age_group, gender, pattern, and size_type improve matching for filtered and specific searches, and some are required in apparel. Fill every attribute that applies. Completeness is a ranking-relevant signal in its own right.
• Supplemental feeds: a second feed that adds to or overrides fields in your primary feed, joined by item ID. This is the practical way to layer optimizations, such as custom labels, better titles, promotion IDs, or fixed categories, without rebuilding the primary feed your platform generates. It is the workhorse of ongoing feed optimization, because it lets you iterate fast on top of a stable base.

For stores built on Shopify, Salla, or Zid, the primary feed usually comes straight from the platform, and a supplemental feed is where a specialist adds the enrichment the platform does not produce on its own.

Frequently Asked Questions

What is product feed optimization in Google Merchant Center?

It is the ongoing work of structuring and enriching your product data, titles, descriptions, categories, identifiers, images, prices, and labels, so that Google matches your products to the right searches and your Shopping and Performance Max campaigns earn more qualified clicks at a better return on ad spend. Because these campaigns have no keywords, the feed is what Google reads to decide where your products appear.

How long should a product title be?

The limit is 150 characters. Front-load the most important keywords into the first 70 characters, since that is roughly what displays before truncation in many placements. Follow a category formula, typically Brand plus Product Type plus Attributes plus Color and Size, and avoid any promotional words, which are disallowed in titles.

Do I really need GTINs on every product?

Provide a GTIN whenever the product has one, which covers almost all branded and manufactured goods. Correct GTINs are among the strongest matching signals in the feed and are effectively expected for those items. Only set identifierExists to "no" for genuinely custom, handmade, or own-brand products that have no barcode or manufacturer part number.

Can I put a "sale" badge or my logo on the product image?

No. The main image must show the product clearly, usually on a plain white background, with no promotional overlays, badges, watermarks, borders, or call-to-action text. These cause disapproval. To show a discount, use the sale_price field, which lets Google display an honest strike-through price without any overlay.

What are custom labels used for?

Custom labels are five self-defined tags that shoppers never see. You use them to segment campaigns by margin, bestseller status, season, price band, or performance tier, so you can bid, budget, and report on each group separately. They turn one undifferentiated catalog into groups you can steer independently.

How does a better feed improve return on ad spend?

A precise, keyword-rich, accurate feed shifts spend away from generic low-intent matches toward qualified searches that convert. Strong titles and complete attributes lift impressions and click-through rate on the right queries, accurate identifiers and categories tighten matching, and clean images and honest sale pricing raise the click rate. Practitioners commonly see improved impressions, click-through rate, and return on ad spend after this work, with the size of the lift varying by catalog.

Related

• Google Shopping ads management in Dubai
• Performance Max campaigns in Dubai
• Ecommerce PPC in Dubai
• Ecommerce development and store builds
• Shopify marketing in Dubai
• Pricing
• Contact Big AL