North American Media Experts
Learn how to set up Google Shopping ads with a proven 6-step campaign structure — from Merchant Center setup and feed optimization to bidding strategy and search term hygiene.
9 min read
By the end of this guide, you will have a fully structured Google Shopping campaign that matches your products to high-intent buyers, controls your spend at the product level, and gives Google's algorithm the data it needs to improve your return on ad spend over time. Google Shopping ads consistently outperform text ads in e-commerce — average ROAS for Shopping campaigns sits at 4.1x — but that number only materializes when the campaign structure is right from the start.
Setting up Google Shopping ads is not just clicking a button and uploading a product list. There are six distinct layers — Merchant Center, feed quality, campaign type, campaign structure, bidding, and negative keywords — and each one affects your performance. Miss one and you are essentially paying for a traffic source with a slow leak.
Google Shopping runs through Google Merchant Center, not directly through Google Ads. Your first task is creating a Merchant Center account (merchant.google.com), verifying ownership of your domain, and then linking that account to your Google Ads account. This link is what allows your product data to flow into campaigns.
Once linked, submit your product feed. A product feed is a structured file — typically in XML or TXT format via a Google Sheet, a direct URL, or a scheduled fetch — that contains every product you want to advertise. At minimum, each product entry needs:
Merchant Center will flag feed errors immediately. Resolve every error and warning before running campaigns — disapproved items simply will not show. Pay particular attention to price mismatches (Google crawls your site to verify) and missing GTINs for branded products, which reduce your auction eligibility.
Your product title is your primary keyword signal in Google Shopping. Unlike Search ads where you write explicit keywords, Shopping matches your products to queries based primarily on your title and description. This means feed quality is your creative strategy — the highest-leverage optimization in the entire channel.
Google allows up to 150 characters in titles, but only the first 70 are visible in most Shopping ad formats. Front-load the most important terms. The proven structure for most product categories is:
Brand → Product Type → Key Attribute(s) → Size/Color/Variant
For example, instead of "Men's Running Shoe — Lightweight Style", write "Nike Air Zoom Pegasus 41 Men's Running Shoe — Lightweight, Size 10, Grey". The second version matches queries like "Nike Pegasus 41 grey size 10" and "men's lightweight running shoe Nike" — two completely different search intents, both captured.
According to feed optimization data from 2026, optimized titles increase impressions by 15–30% and improve CTR by 10–20% on their own. This single change often produces the largest ROAS improvement of any Shopping optimization.
Beyond titles, fill in every optional attribute you have data for: product_type (your own taxonomy, used for campaign segmentation), custom_label_0 through custom_label_4 (your own tags, like "high-margin", "seasonal", "clearance"), and google_product_category (Google's taxonomy). Custom labels are especially useful — they let you segment campaigns by profit margin rather than just product type, which changes how you bid entirely.
In 2026, you have two options for Google Shopping campaigns: Standard Shopping and Performance Max. Each has a role.
Standard Shopping gives you explicit control: you bid at the product group level, you run your own negative keyword lists, and you can see search term reports at a granular level. It is lower reach but higher transparency. Average conversion rate for Standard Shopping is around 1.9%.
Performance Max (PMax) uses Google's machine learning to show your products across Search, Shopping, YouTube, Display, Gmail, and Maps simultaneously. It averages a 2.4% conversion rate and 4.1x ROAS for e-commerce accounts with sufficient data. For accounts with 50 or more conversions per month, PMax Shopping typically outperforms Standard Shopping by 15–20% on ROAS — but the trade-off is less visibility into where your spend goes.
The recommended approach for most brands: start with Standard Shopping to build conversion history for 60–90 days, then layer in a Performance Max campaign targeting the same products. Use asset groups in PMax to mirror your product categories. This is also worth reading alongside our guide to structuring a Performance Max campaign from scratch, which covers the full 7-step framework.
Do not run Standard Shopping and PMax on the exact same products without a priority system — they will cannibalize each other. Either use PMax alone once you have data, or use campaign priority settings to control which campaign handles which queries.
One of the most common mistakes in Google Shopping is treating all products identically. A $500 margin item and a $5 margin item should not have the same target ROAS or the same budget allocation.
Use a three-tier campaign structure based on your custom labels:
Within each campaign, build product groups by custom_label (margin tier, season, or clearance status) rather than by category alone. This gives you bid control at a level that actually maps to your business economics. If your custom_label_0 is "high-margin", you can bid 30–40% higher for those products and still hit your target return.
Running a regular review of your campaign structure is also part of healthy account hygiene — our 10-step Google Ads account audit checklist includes specific Shopping checks you should run monthly.
For new Shopping campaigns with limited conversion history, start with Maximize Clicks to build data, then transition to Target ROAS once you have at least 30–50 conversions recorded. Target ROAS tells Google how much revenue you want for every dollar spent — a target of 400% means you want $4 back for every $1 in ad spend.
Setting your ROAS target too aggressively too early starves the campaign of traffic before the algorithm has enough data. A common approach is to set your initial ROAS target at 20–30% below your actual goal, let the campaign hit volume, then tighten it over 2–3 weeks.
Budget allocation: as a general baseline, allocate roughly 60% of your Shopping budget to your highest-margin products and 40% to broader category campaigns. Rebalance monthly based on actual ROAS by product group.
For a deeper dive into bidding options across Google's auction ecosystem, see our guide to choosing the right Google Ads bidding strategy in 2026, which walks through every automated and manual option with decision criteria for each.
Google Shopping does not use traditional keyword targeting, but it absolutely uses negative keywords to exclude irrelevant queries. Without negatives, your product ads will appear for searches like "DIY [product name]", "[product name] review", or competitor brand names that share terminology with yours — all of which drain budget without converting.
Build your negative keyword lists in layers:
Review your search term report weekly for the first 90 days and monthly thereafter. Look for any query spending more than 3x your target CPA with no conversion — add it as a negative immediately.
Google Shopping has its own set of KPIs. The ones that matter most are:
Impression Share (IS) — what percentage of eligible auctions you actually appeared in. If your IS is below 60%, you are either being outbid or your budget is too low. "Lost IS (rank)" tells you which it is.
Click-through rate (CTR) — Shopping CTR benchmarks range from 0.7% (broad categories, low relevance) to 2.5%+ (strong brand + high-intent queries). Low CTR usually signals a weak product title or a poor image.
Conversion rate by product group — a Shopping-wide average masks the difference between a 5% converter and a 0.5% converter in the same campaign. Segment by product group in your reports.
ROAS by custom label — this tells you whether your margin segmentation is working. If your "high-margin" label is hitting 6x ROAS and your "clearance" label is at 2x, your structure is doing its job.
Even well-intentioned setups fail for predictable reasons:
Ignoring feed quality after launch. Your feed needs to be refreshed at minimum daily (ideally every 6 hours for price-sensitive categories). Stale prices or out-of-stock items that still appear in ads lead to wasted spend and policy violations.
Setting ROAS targets before there is data. Target ROAS needs conversion history. Launching with a 500% ROAS target on day one will cause the campaign to barely serve.
Treating Shopping as a set-it-and-forget-it channel. Shopping campaigns degrade over time if not maintained. New competitors enter the auction, seasonal trends shift, and product assortment changes. Monthly reviews are the minimum.
Not segmenting by margin. Uniform bidding across a product catalogue is one of the most common ways e-commerce brands leave money on the table. Your highest-margin items should almost always get higher bids and more budget.
Skipping the search term audit. This is non-negotiable in the first 90 days. Without it, irrelevant queries will quietly consume 20–30% of your budget.
Setting up Google Shopping correctly from the start — Merchant Center verified, feed optimized, campaigns structured by margin, bidding matched to your conversion history, and negatives in place — is what separates accounts that hit 5–8x ROAS from those stuck at 2x. The channel rewards structure and ongoing attention, and the brands that get it right early build a compounding advantage as their conversion data improves Google's targeting over time.
If you want a team to build and manage your Google Shopping campaigns — or audit an existing setup that is not delivering — our paid search specialists work with e-commerce brands across North America to maximize Shopping performance. Get a quote and let us take a look at your current setup.