Product feed optimization: titles, attributes, images
Feed optimization is how you win more shopping traffic without raising bids. Optimize titles, attributes, and images to improve match rate and ranking.
Why feed optimization beats bidding
Shopping channels do not have keywords you bid on the way text ads do. Instead, the channel reads your feed and decides which searches your product is relevant to. That means the words and attributes in your feed are your targeting. Optimizing the feed is the highest-leverage thing you can do, because it improves which queries you show up for and how well you rank, before you spend a cent more.
A well-optimized feed gets matched to more relevant searches, earns higher click-through rates, and avoids the disapprovals that quietly drain your catalog. Here is where to focus.
Titles: your most important attribute
The title is the single biggest driver of shopping performance. The channel weighs it heavily when matching your product to searches, and shoppers read it before they read anything else. Most stores ship titles that are too short, too vague, or stuffed with internal jargon.
Lead with the attributes shoppers actually search for. A strong title structure puts the most important details first because titles get truncated. A common, effective pattern is brand, then product type, then key attributes like color, size, material, and model.
- Weak: TrailRunner Shoe
- Strong: Acme TrailRunner Mens Running Shoes, Red, Size 9, Lightweight Mesh
- Put high-value keywords in the first 70 characters; aim to use the available length.
- Match how people search: include color, size, material, gender, and model number.
- Avoid promotional text in titles (Sale, Free Shipping, Best). Channels disapprove it.
Attributes: completeness wins
Channels reward complete data. The more required and recommended attributes you fill in correctly, the more contexts your product can appear in and the more likely it is to be approved. Missing GTINs, empty brand fields, and absent product categories are the most common gaps, and each one costs you visibility.
Prioritize the attributes that drive matching and eligibility: GTIN and brand (which connect your product to the global catalog), google_product_category (which defines what you are selling), product_type (your own taxonomy), and the variant attributes color, size, gender, age_group, and material. For apparel especially, the variant attributes are not optional if you want strong performance.
- Fill GTIN, brand, and MPN wherever they exist to improve match rate.
- Set google_product_category accurately; a wrong category mismatches your product.
- Populate color, size, material, gender, and age_group for apparel and variants.
- Use product_type for your own navigation taxonomy alongside the Google category.
Images: meet the rules, then stand out
Image requirements are strict and disapprovals here are common. The main image generally must show the product on a plain background, be large enough (Google asks for a minimum size and prefers high resolution), and contain no promotional text, watermarks, logos, or borders overlaid on it. Break those rules and the product gets disapproved.
There is an important nuance for ads. Promotional overlays like a discount badge are not allowed on Google Shopping product images, but Google offers sale price and promotion features for that purpose. On social channels like Meta, TikTok, and Snapchat, creative overlays are widely used and can lift performance, so you often want the same product with a clean image for Google and a badged image for social.
MartechFlow handles both sides of this. It validates images against each channel's rules and, with its built-in image creative, can render a clean image for Google and a badged version with a live discount for social channels, computed straight from your feed's sale and original prices. You optimize once and each channel gets a compliant, channel-appropriate image.
Make optimization a rule, not a chore
Optimizing 50 products by hand is doable. Optimizing 50,000 is not, and it would break the moment your catalog changes. The point of feed optimization is to express it as repeatable rules: build titles from a template, fill missing categories from a lookup, append variant attributes automatically, and badge images from price logic.
In MartechFlow you set these rules once with a simple interface, and they apply to your whole catalog on every refresh. New products inherit the same optimized structure automatically, so your feed stays clean as you grow without anyone re-editing it.