eCommerce Search Campaigns: Structure for Product-Level Bidding

eCommerce Search campaigns get squeezed between Shopping (which covers most product queries) and PMax (which covers Search + Display + Video automatically). But there are 3 specific query patterns where dedicated Search campaigns still win - brand queries, exact-product queries, and category-plus-modifier queries.

In the modern Google Ads ecosystem, the role of Search has shifted from being the primary driver of traffic to being the precision tool in a wider automated arsenal. While Performance Max (PMax) and Standard Shopping do the heavy lifting of visual discovery, Search campaigns remain the only place where you have absolute control over the text the user sees, the landing page they hit, and the specific USP you highlight for their exact intent.

At Visionary Marketing, we've seen too many accounts abandon Search entirely in favour of "all-in-one" automation. This is a strategic error that leaves high-intent revenue on the table for competitors. This guide details the SKAG-lite structure we deploy for eCommerce Search, DSA integration for large catalogues, and the negative-keyword strategy that keeps Search + Shopping from cannibalising each other.

1. Where Search still wins vs Shopping + PMax

The "Shopping Squeeze" is real. When a user searches for a product, Google prefers to show a visual grid of products. If you are bidding on "blue running shoes", your text ad is competing with a dozen photos of blue running shoes. In that battle, the photo wins 9 times out of 10.

However, there are three patterns where Search text ads outperform or essentially supplement the Shopping experience to secure the click:

A. Brand Queries

Brand queries are the most valuable traffic you have. While PMax will bid on your brand, it does so with a generic "catch-all" approach. A dedicated Brand Search campaign allows you to:

• Message Control: Highlight specific current offers (e.g., "Extra 10% off for Members") that aren't in your main site metadata.
• Sitelink Strategy: Force users toward high-margin collections or new arrivals rather than just the homepage.
• Competitive Defence: Prevent competitors from "conquesting" your brand name with superior ad copy.

Even if you use PMax brand exclusions, you still need a home for this traffic. A dedicated Search campaign is that home.

B. Exact Product-Name Queries

When a user searches for "Nike Pegasus 40 FlyEase", they are at the very bottom of the funnel. A Shopping card shows them the shoe. A Search ad above it can say: "Nike Pegasus 40 FlyEase - Free Next Day Delivery - Official Stockist".

The text ad addresses the utility of the purchase (delivery speed, trust) whereas the Shopping card addresses the visual. In many cases, the user clicks both, but having the top text ad increases your "SERP real estate" and pushes competitors further down the page.

C. Category-plus-Modifier Queries

These are queries like "best running shoes for flat feet" or "cheap dishwashers under £300". Shopping is terrible at modifiers because it relies on feed attributes which are often incomplete. Search, however, allows you to create an ad group specifically for "flat feet" and write ad copy that says "Engineered for Flat Feet - Shop the Collection". This relevance boost is a massive CTR driver.

2. The SKAG-lite structure for eCommerce Search

Years ago, the gold standard was SKAGs (Single Keyword Ad Groups). You would have one ad group for [red nike shoe], one for [blue nike shoe], etc. In the era of "Close Variants," where Google treats "running shoes" and "jogging sneakers" as the same thing, traditional SKAGs are dead. They create too much data fragmentation for Smart Bidding to work.

SKAG-lite is our modern evolution. Instead of one keyword per ad group, we use one product family per ad group.

• The Keywords: 5-10 keywords that are variations of the same specific intent. For example: [product name], "product name", +product +name +reviews.
• The Ads: 3 Responsive Search Ads (RSAs) where the headlines are dynamically inserted but the descriptions focus on the specific benefits of that product family.
• The Bidding: Because you have more than one keyword, you get enough conversion data per ad group to feed the algorithm, which isn't possible with traditional SKAGs.

The goal of SKAG-lite is manageable relevance. You want enough ad groups to be relevant, but not so many that you spend 40 hours a week just checking budgets. We typically limit SKAG-lite treatment to the top 20% of products that drive 80% of revenue. The rest? They go to DSA.

3. Brand-only campaign setup

A Brand Search campaign is not a "set it and forget it" task. It is your most profitable campaign and should be treated with the highest level of detail. We recommend a two-ad-group structure:

The Bidding Decision: When should you use "Target Impression Share" vs "Max Conversions"? If you are a market leader being attacked by competitors, use Target Impression Share (95%+) to ensure you always own the top spot. If you are a smaller brand, Max Conversions is usually better as it allows Google to bid more aggressively when the user's signals suggest a high likelihood of purchase.

One contrarian point: we often see "Brand" ROI look so good that it masks poor performance elsewhere. Always view your account performance *excluding* brand to understand your true growth engine. However, do not stop bidding on brand. The moment you do, your competitors will bid £0.10 for your name and steal your most loyal customers.

4. Product-name campaign structure

Product-name campaigns are the bridge between "I'm looking for a shoe" and "I'm buying THIS shoe." For a Shopify store with 2,000 SKUs, you cannot have 2,000 ad groups.

Our workflow for Product-Name campaigns:

1. Export Top 100 Revenue SKUs: Use GA4 or your internal sales data.
2. Create "High Intent" Ad Groups: One per SKU from that top 100.
3. Implement Dynamic Ad Customisers: Use Google's business data feeds to automatically pull the current price or stock level into the ad headline. Nothing converts better than "Buy [Product] - Only £49.99 - In Stock" when the price is accurate.
4. The "Tail" Catch-all: All other SKUs are handled by a high-efficiency DSA campaign (see Section 6).

This hybrid approach ensures that your money is spent where the impact is highest. You don't need a manual ad for a £5 replacement shoelace, but you definitely need one for the £150 running shoe.

5. Category-plus-modifier campaigns

This is where you capture the "pre-shopping" intent. People searching for "waterproof winter coats" haven't decided on a brand yet.

The mistake most agencies make is sending this traffic to the homepage or a generic "Coats" collection. To win here, you need to match the modifier in the ad copy and the landing page:

• Modifier: "Cheap" > Ad Copy: "Budget Friendly Prices" > Landing Page: Sorted by price low-to-high.
• Modifier: "Best" > Ad Copy: "Top-Rated & Award-Winning" > Landing Page: Sorted by reviews.
• Modifier: "Sustainable" > Ad Copy: "Eco-Friendly Materials" > Landing Page: Pre-filtered for "Recycled".

This level of granularity is what separates a 2.0x ROAS account from a 6.0x ROAS account. It's about reducing the "cognitive load" for the user. If they ask for "sustainable," show them "sustainable" immediately.

6. DSA integration for large catalogues

Dynamic Search Ads (DSA) are often misunderstood as "lazy marketing." In reality, they are an essential data-gathering and gap-filling tool. Google's crawler is better at matching long-tail queries than your keyword research will ever be.

For eCommerce, we recommend **Page Feed DSA**. Instead of letting Google crawl your whole site, you provide a CSV of URLs (from your Shopify/WooCommerce product export). This prevents Google from bidding on your blog posts, shipping policy, or contact page.

DSA is also your safety net. If you forget to build a campaign for a new product launch, DSA will pick up the slack within hours of the URL being live. It is the insurance policy of a modern eCommerce account. Read our full DSA guide for more.

7. Preventing Search + Shopping cannibalisation

If you don't manage your negative keyword lists, your Search and Shopping campaigns will compete for the same user. This doesn't necessarily mean you'll pay twice for the same click (Google's auction prevents that), but it does mean you lose control over which ad format serves.

If a user searches for your brand, you want them to see the Search ad with sitelinks. If you don't exclude your brand from Shopping, Google might show a Shopping tile instead. The ROI might still be okay, but you've lost the chance to show them your "Best Sellers" sitelink or "Summer Sale" headline.

We recommend using **Shared Negative Keyword Lists**. Create one list called "Shopping Exclusions" and apply it to your Shopping campaigns. Create another called "Search Exclusions" and apply it to Search. This makes management significantly easier as you only have to add a keyword once to update the entire account.

8. Bidding strategy for Search campaigns

eCommerce Search bidding is not a monolith. You should vary your strategy based on the intent of the campaign:

Always wait for a campaign to have at least 30 conversions in a 30-day period before switching from "Max Conversions" to "Target ROAS". The algorithm needs that data to understand what a "good" user looks like.

9. Common eCommerce Search mistakes

In our audits of UK eCommerce accounts, we see the same five mistakes repeated by internal teams and generalist agencies:

Frequently asked questions