Google Shopping Campaigns: What Actually Drives Revenue

A Google Shopping commercial, more precisely called a Shopping campaign or Product Listing Ad (PLA), is a paid placement that shows your product image, price, and retailer name directly in Google search results. Unlike text ads, Shopping ads pull from a product feed rather than keywords you manually bid on, which changes how you build, manage, and optimise them considerably.

If you sell physical products online and you are not running Shopping campaigns, you are almost certainly leaving revenue on the table. The format converts well because it shows the product before the click, which pre-qualifies the buyer in a way that text ads simply cannot match.

Shopping campaigns sit within a broader paid advertising ecosystem that most businesses underestimate in complexity. If you want the full picture of how paid channels fit together, the Paid Advertising Master Hub covers the strategic layer that most platform-specific guides skip entirely.

How Google Shopping Ads Actually Work

The mechanics are worth understanding properly because they shape every decision you make downstream. When someone types a product-related query into Google, the Shopping algorithm matches it against product feeds submitted through Google Merchant Center. It then decides which products to show, in which order, at what price point, based on a combination of bid, feed quality, and relevance signals.

There are no keywords to bid on in the traditional sense. You are not writing ad copy. What you are doing is structuring a product feed that Google reads and interprets. This is a fundamentally different skill set from running search text ads, and it catches a lot of advertisers off guard when they first make the transition.

The feed itself contains product titles, descriptions, images, prices, availability, GTINs (Global Trade Item Numbers), and a range of optional attributes that can improve matching. Google uses the product title and description as the primary signals for determining which queries your products appear against. This is why feed optimisation is not a nice-to-have. It is the core of the work.

I have seen advertisers spend months adjusting bids and restructuring campaigns while leaving product titles that read like internal SKU codes. The titles were things like “BLK-PREM-42-M” instead of “Men’s Black Premium Running Shoes, Size 42.” The targeting was a mess, not because of campaign structure, but because the feed was telling Google almost nothing useful. Fix the feed first. Everything else is secondary.

Google Shopping has evolved considerably since its early days as Froogle, a free product search tool. The transition to a paid model integrated with AdWords changed the competitive dynamics entirely and set the stage for the auction-based system that exists today.

Standard Shopping vs. Performance Max: Choosing Your Campaign Type

Google currently offers two primary campaign types for Shopping: standard Shopping campaigns and Performance Max. Understanding the difference matters because they represent genuinely different philosophies about who controls the campaign.

Standard Shopping campaigns give you control over product groups, bids, negative keywords, and device adjustments. You can see search term reports (with some limitations post-2020). You can segment products into groups and bid differently on high-margin versus low-margin SKUs. The transparency is relatively good. The manual work is real.

Performance Max, which Google pushed aggressively from 2021 onward, consolidates Shopping, Display, YouTube, Gmail, and Discover into a single campaign. Google’s algorithm decides where to show your ads, who to show them to, and at what bid. You provide asset groups (images, headlines, descriptions, logos) and a product feed, and the machine does the rest.

The pitch from Google is efficiency at scale. The reality is a significant reduction in visibility into what is actually happening. You lose granular search term data. You cannot easily isolate Shopping performance from Display performance. Budget attribution becomes murky. For advertisers managing significant spend across multiple product categories, that opacity is a real problem.

My view, shaped by managing hundreds of millions in ad spend across thirty industries, is that Performance Max works reasonably well when your feed is strong, your conversion tracking is clean, and you have enough historical data for the algorithm to learn from. It underperforms when any of those conditions are missing, and it does so quietly, which is the dangerous part. You will not always see the signal that something has gone wrong.

For businesses that want to understand the full range of campaign management decisions involved in paid search, PPC management services covers the operational and strategic considerations in more depth.

Feed Quality: The Variable Most Advertisers Undervalue

If I had to identify the single biggest performance gap I have seen across retail advertisers running Shopping campaigns, it would be feed quality. Not bids. Not campaign structure. The feed.

A product feed that is accurate, complete, and well-structured does several things simultaneously. It improves the relevance of query matching. It reduces wasted spend on irrelevant searches. It increases the likelihood of appearing in the right placements at the right price point. And it feeds better data into Google’s machine learning, which compounds over time.

Product titles are the most important element. Google uses them as the primary matching signal. A good title follows a structure that front-loads the most important attributes: brand, product type, key differentiators (colour, size, material), and model number where relevant. For a women’s running shoe, that might look like: “Nike Women’s Air Zoom Pegasus 40, Grey/White, UK Size 6.” That title tells Google everything it needs to match the product to relevant queries.

Descriptions matter less for matching but still influence quality scores and can affect click-through rates in some placements. Images are critical for click-through rate. A product image on a clean white background typically outperforms lifestyle imagery in Shopping placements, though this varies by category.

GTINs (barcodes, ISBNs, MPNs) help Google understand exactly what product you are selling and match it against manufacturer data. Missing GTINs for products that have them is a common feed gap that limits eligibility for certain placements.

Price and availability data must be accurate and updated regularly. Google cross-references your feed against your landing page. Discrepancies cause disapprovals, which pull products out of auction entirely. I have seen campaigns lose 20-30% of their eligible inventory to disapprovals that no one had noticed because the feed management process was not strong enough.

Campaign Structure and Product Segmentation

Once your feed is in good shape, campaign structure becomes the next lever. The goal of structuring a Shopping campaign is to give yourself the ability to bid differently on products that have different commercial characteristics.

Not all products deserve the same bid. A product with a 60% margin deserves a higher target CPA than a product with a 15% margin. A bestseller with strong conversion data deserves different treatment than a new product with no history. A product you are clearing at cost should not compete for budget with your hero SKUs.

The standard approach is to segment products into groups based on category, margin tier, or performance tier, and set different target ROAS (Return on Ad Spend) goals for each. This sounds straightforward. In practice, it requires clean data on margin by SKU, which many retailers do not have readily accessible, and it requires a feed structure that allows you to create the segments you need.

Custom labels in the feed are the mechanism for this. You can apply custom labels to products (up to five per product) to tag them with attributes that do not exist in the standard feed specification. Common uses include margin tier (high, medium, low), seasonality (clearance, new season), or performance classification (hero, standard, long-tail). These labels then become the basis for product group segmentation inside your campaign.

Priority settings matter in standard Shopping campaigns. When the same product appears in multiple campaigns, priority settings determine which campaign enters the auction. This is how you build campaign structures that funnel high-intent queries to campaigns with higher bids while letting lower-priority campaigns catch broader traffic at lower bids. It is a useful technique, though it adds structural complexity that needs ongoing maintenance.

Bidding Strategy: Automation With Eyes Open

Google’s automated bidding strategies have improved meaningfully over the past several years. Target ROAS and Target CPA bidding can work well when you have sufficient conversion volume (Google recommends at least 30-50 conversions per month at the campaign level, and more is better) and when your conversion tracking is accurate.

The failure mode I see most often is advertisers applying automated bidding to campaigns with thin conversion data. The algorithm does not have enough signal to optimise effectively, so it makes poor decisions, performance deteriorates, and the advertiser either abandons the strategy or blames the platform. The real issue was applying automation before the conditions for it were met.

Manual CPC bidding still has a place, particularly for new campaigns where you are still building data, or for advertisers who want granular control over spend by product. It requires more active management but gives you a clearer line of sight into where money is going.

One thing I have learned from running campaigns across dozens of categories is that the target ROAS you set is not just a performance goal. It is a constraint on volume. Set it too high and the algorithm restricts spend to only the most efficient conversions, which may mean you are leaving profitable volume on the table. Set it too low and you are buying volume at a loss. Finding the right balance requires understanding your unit economics, not just your platform metrics.

This connects to a broader point about Google Ads (formerly Google AdWords) as a platform: the interface surfaces a lot of data, but the commercially relevant question is always whether the margin after ad spend is positive, not whether your ROAS looks impressive in a dashboard.

Google Shopping for Specific Verticals

Shopping campaigns are most obviously suited to e-commerce retailers selling physical products. But the format has relevance across a broader range of verticals than many advertisers realise.

Fashion and apparel is one of the highest-volume categories on Google Shopping. The visual nature of the format suits it well. Margin structures in fashion vary enormously, which makes proper product segmentation by margin tier especially important. Seasonal inventory management, where you need to adjust bids as stock levels change, adds another layer of complexity.

Health and beauty is another strong vertical for Shopping. Consumers searching for specific products, brands, or formulations convert well because the intent is typically high and specific. If you are running paid acquisition for a beauty or wellness brand, Google Ads for beauty businesses covers the vertical-specific considerations worth knowing.

Home and garden, electronics, sporting goods, and pet supplies are all categories where Shopping performs consistently well. The common thread is product specificity: when a consumer knows roughly what they want and is in a comparison or purchase mindset, Shopping ads are highly effective at intercepting that demand.

What Shopping is less suited to is genuinely complex or considered purchases where the decision involves significant research, customisation, or consultation. High-value B2B products, bespoke services, and products that require significant explanation before purchase tend to convert poorly from Shopping placements. The format is built for the moment when someone is ready to buy something they already understand.

The Cost Structure Behind Shopping Campaigns

Shopping campaigns operate on a cost-per-click model. You pay when someone clicks your ad, not when it is shown. The cost per click varies by category, competition, and product type. Highly competitive categories like consumer electronics or branded fashion can see CPCs that make the economics challenging unless your conversion rate and average order value are strong.

Understanding the full cost structure of running Shopping campaigns, including platform fees, agency fees if applicable, and the relationship between CPC and ROAS targets, is important before you commit significant budget. Google advertising fees breaks down the cost components in a way that helps you model the economics before you are committed to a spend level.

One calculation that is worth doing before you set your ROAS target is working backwards from margin. If your average order value is £80 and your gross margin is 40%, you have £32 of margin to work with before you account for fulfilment, returns, and customer service costs. If your net margin after those costs is £20, you need your ad spend to be below £20 per order to be profitable. That translates to a minimum ROAS of 4x. Set your target below that and you are subsidising sales with margin. It sounds obvious. You would be surprised how often it gets skipped.

Early in my career at lastminute.com, I ran a paid search campaign for a music festival that generated six figures of revenue within roughly 24 hours. It was a relatively simple campaign, but the economics were right: the margin on the ticket inventory was strong, the demand was there, and the targeting was clean. The lesson was not that paid search is magic. It was that when the unit economics work and the demand exists, paid search is one of the fastest ways to capture it. Shopping campaigns operate on the same logic.

Negative Keywords and Query Management

Because Shopping campaigns do not use keyword lists in the traditional sense, negative keywords take on extra importance. They are the primary tool for excluding irrelevant traffic and tightening the match between your products and the queries that trigger them.

A strong negative keyword strategy for Shopping typically involves several layers. Brand negatives prevent your products from showing for competitor brand queries where you are unlikely to convert. Category negatives exclude product types you do not sell. Intent negatives filter out informational queries from people who are researching rather than buying.

Search term reports (available in standard Shopping campaigns, limited in Performance Max) are the primary source of data for building and refining negative lists. Reviewing these regularly, particularly in the early weeks of a campaign, is essential. I have seen Shopping campaigns where 30-40% of spend was going to queries that had no reasonable chance of converting, simply because no one had reviewed the search terms and added negatives.

In Performance Max, the lack of granular search term visibility makes this harder. You can add negative keywords at the account level or request campaign-level negatives through Google support, but the process is less transparent and less controllable than in standard Shopping. This is one of the practical reasons why some advertisers maintain standard Shopping campaigns alongside Performance Max rather than migrating entirely.

Measurement, Attribution, and the Honest Approximation

Shopping campaign measurement sits within the broader attribution problem that affects all paid channels. Google’s own attribution models favour Google. Last-click attribution understates the value of upper-funnel activity. Data-driven attribution is better but still opaque. Cross-device journeys are partially invisible. None of this is unique to Shopping, but it matters when you are making budget decisions.

The practical approach is to treat platform-reported ROAS as one signal among several rather than the definitive measure of performance. Triangulate it against revenue data in your analytics platform, against order data in your e-commerce system, and against any incrementality testing you can run. When they all point in the same direction, you can be reasonably confident. When they diverge, you need to understand why before you make major budget decisions.

Conversion tracking accuracy is non-negotiable. If your tracking is misconfigured and you are double-counting conversions, your ROAS will look better than it is and you will overbid. If you are missing conversions, your ROAS will look worse and you will underbid. Both errors are costly. Auditing conversion tracking before scaling spend is not optional.

I walked into a CEO role once and spent my first weeks scrutinising the P&L rather than the marketing dashboards. The marketing team had impressive-looking performance metrics. The P&L told a different story. I told the board the business would lose around £1 million that year. That is almost exactly what happened. The lesson applies to Shopping campaigns: the metrics in the platform interface are not the same as commercial reality. Always trace the numbers back to margin.

It is also worth considering how Shopping fits within your broader channel mix. If you are running TikTok Ads or other upper-funnel channels alongside Shopping, the demand that Shopping captures may partly be demand that your other channels created. Shopping is primarily a demand capture channel. It converts people who are already in a buying mindset. If you only run Shopping and nothing else, your growth ceiling is set by the organic demand that already exists for your products.

When to Manage Shopping In-House vs. Using an Agency

This is a question I get asked often, and the honest answer is that it depends on the complexity of your feed, the scale of your spend, and the internal capability you have available.

Shopping campaigns are technically accessible. The interface is not complicated. But managing them well at scale, with proper feed optimisation, segmentation, bidding strategy, and ongoing negative keyword management, requires time and expertise. A small retailer with a clean feed and a simple product range can manage Shopping reasonably well in-house with some initial guidance. A retailer with thousands of SKUs across multiple categories, complex margin structures, and significant spend needs either a strong in-house specialist or external support.

If you are evaluating external support, understanding what a PPC agency actually does and how to evaluate one is worth doing before you start conversations. The questions you should be asking are about feed management capability, how they handle product segmentation, and what their reporting looks like beyond platform metrics.

One thing I would caution against is treating Shopping campaign management as a commodity. The difference between a well-managed and a poorly managed Shopping account at scale can be significant in terms of wasted spend and missed revenue. The platform makes it look simple. The commercial complexity underneath is real.

AI tools are increasingly being applied to Shopping campaign management, particularly for feed optimisation and bid management. Using AI to improve Google Ads campaigns is a genuinely useful application when the fundamentals are already in place. It is not a substitute for understanding the underlying mechanics.

If you are building out your paid acquisition capability more broadly, the Paid Advertising Master Hub covers the strategic and operational questions that sit above any individual channel, including how to think about channel mix, budget allocation, and measurement across a portfolio of paid activity.

What Good Shopping Performance Actually Looks Like

There is no universal benchmark for Shopping ROAS that applies across categories. A 4x ROAS might be excellent for a low-margin electronics retailer and inadequate for a high-margin accessories brand. The right benchmark is the one derived from your own unit economics, not an industry average pulled from a platform blog post.

What good Shopping management looks like in practice is a combination of: a feed that is accurate, complete, and regularly updated; campaign structure that reflects the commercial differences between product groups; bidding strategy calibrated to margin targets rather than vanity ROAS; a negative keyword list that is actively maintained; and measurement that goes beyond platform reporting to connect ad spend to actual margin.

The businesses I have seen get the most from Shopping campaigns are the ones that treat the feed as a living asset rather than a one-time setup task, that have clean conversion tracking, and that have someone who understands both the platform mechanics and the commercial context well enough to make good decisions when the data is ambiguous.

Shopping campaigns are not complicated in concept. They are demanding in execution. The gap between a campaign that works and one that wastes budget is usually found in the details: a feed that has not been updated in six months, a ROAS target set without reference to margin, a negative keyword list that was never built. Those are fixable problems. But you have to be looking for them.

About the Author

Keith Lacy is a marketing strategist and former agency CEO with 20+ years of experience across agency leadership, performance marketing, and commercial strategy. He writes The Marketing Juice to cut through the noise and share what works.

Frequently Asked Questions