Google Shopping has become the dominant force in retail advertising.
Shopping ads account for an estimated 76% of retail search ad spend and generate roughly 85% of all clicks on paid search campaigns.
Online retailers who aren’t leveraging this network as part of their media strategy may be leaving substantial revenue on the table (context aside).
Shopping campaigns have been proven to deliver higher efficiency than traditional text ads with studies showing that Shopping ads drive ~30% higher conversion rates than text ads.
However, we often hear e‑commerce teams say “we tried Shopping ads and they didn’t work” , or, “we tried shopping and it just didn’t quite achieve what performance max could for us”.
In many cases, the culprit isn’t the channel itself but execution.
Issues like poor product data, messy campaign structuring, or passive bidding strategies, amongst others.
To really scale Shopping campaigns, you need to get the fundamentals right and then layer on advanced tactics.
And with all things Google Ads related, even with everything in place and a watertight strategy, scaling isn’t just increasing the budget and watching the sales roll in.
It’s about getting this pillar of your media strategy working as best as it can, then stepping out of the accounts to see how this campaign type can be taken further as part of the entire marketing mix.
Below we guide you through the key steps (and best practices) we use to scale Google Shopping for our clients, from data feed optimisation through campaign structure, bidding, and beyond, all backed by industry insights, data and the hands on expertise of the team at outbloom.
Optimise Your Product Data Feed
At the heart of Shopping ads is your product feed, the data file that feeds Google Ads everything it needs to show your products (titles, descriptions, prices, images, and more).
A clean, rich feed is non-negotiable.
In practice, this means:
Complete feed attributes
Every product should have all required fields (ID, title, description, link, image URL, price, etc.) as standard, and this isn’t a set and forget.
Put a process in place to ensure each SKU follows the same structure moving forward.
Additionally, add as many optional attributes as make sense (brand, GTIN/MPN, category, etc.).
The more detail, the better the match with search queries and the higher your chances are of serving your products to the right user at the right time.
Researched titles and descriptions
Craft titles and descriptions that mirror how people search for your products.
Dig into your search query data and feed this into your strategy.
It may be that ‘size’ is super important to your customers (think of furniture retailers selling tables, having size at the front of the title will be helpful with query matching) or with a clothing retailer where the ‘type of fit’ of a t shirt/shirt may be key in getting your ads to serve in more qualified auctions that stand a higher chance of converting.
It goes without saying, you must include key details (color, size, style) and any common synonyms and if you have the data and can map this to your feed, do it.
Google places explicit value in the product title and without the use of keywords, this is where brands can best qualify their traffic (in addition to thorough negative keyword lists).
High-quality product images
Use clean product photos on light backgrounds.
Google advises images fill ~75–90% of the frame, with good lighting and no distracting text or watermarks.
In other words, your main image should clearly show the product in a professional way.
From first-hand experience over the last 15 years, this is often easier said than done.
Think of marketplaces, department stores or electrical component retailers, they may have hundreds of thousands of products from multiple brands and shooting an image of each of these on a white background is just not going to happen.
Be realistic, follow guidelines and always spot check.
Promotions and annotations
If you run sales or promotions, use Googles feed attributes for sale_price or promotion_id to flag discounts and special offers.
Google Ads can then highlight “Sale” call outs on your listings as such:
Don’t forget to add merchant promotions directly in GMC when you go in sale/have ongoing promos (e.g. new customer discounts).
These provide customisable text and a promotional call out, great to message a specific deal. Here’s an example:
Regular updates and quality checks
Setting and forgetting is the biggest mistake we see with product feed driven campaigns.
Stay on top of GMC diagnostics at the very least to spot errors, limitations, etc, then schedule in regular QA across both your key attributes and your custom fields.
Don’t let your feed go stale or assume because it’s got through approvals that it’s fine, there’s always work to be done.
Structure Campaigns For Growth
Once your feed is healthy, you can then move on to building a campaign structure that lends itself to both best practice and your performance data.
Below are some basic rules to follow and there’s not one single structure that can set you up for scale, however, you will need to ensure you have control over your bidding/search queries whilst consolidating performance data.
Segmenting by product groupings or categories
Break out high-priority items into their own campaigns or ad groups.
For instance, create a separate campaign or ad group for top-selling product lines, or for high-margin items.
This allows you to adjust bids and budgets specifically for them without wasted spend on low-margin products.
Separate your campaigns where you have specific KPIs
You might run one campaign for global markets and another for local/regional markets.
Or run a dedicated campaign for new product launches versus your “evergreen” catalog.
You might even create a special campaign for holiday/seasonal merchandise or clearance items, each with their own ROAS targets.
During peak shopping days (Black Friday, Prime Day, etc.), a separate “holiday” campaign lets you raise bids and budgets on those products without affecting your baseline campaigns.
It’s about building a case for the campaign in question, “does this need to be a campaign or can this be an ad group sat within another campaign?”, once you’ve answered this, forecast out how long learnings will take as you scale and be prepared for change.
Campaign priorities
You can use high/medium/low priority to funnel queries: e.g., a high-priority campaign with broad rules to catch all searches, and a low-priority campaign with tight product groupings to catch specific queries.
This trick helps funnel the best matches into your lowest-bid campaigns.
Without keywords, features like this and a well structured feed help to give advertisers control over the auctions they serve their ads in.
A common usage for priorities is to control branded queries that serve on Google Shopping, this involves creating two campaigns, one with a low priority to capture brand another with high priority and your brand negatives to sweep up any generics with lower bids.
Getting this control is key for scaling Google Shopping as you are able to push and pull bids, change budgets, update targeting and more, across key campaign groupings that are prioritised to your goals.
Negative keywords and query funneling
Even though Shopping doesn’t use keywords, you still need to cull irrelevant searches.
Regularly review the Search Terms report and add negatives (e.g. if searches for “used [product]” or unrelated categories are triggering your ads).
This ensures budget isn’t wasted on poor-quality traffic when scaling up spend.
Whether you’re launching new activity or have legacy campaigns, make this a priority as it really is the backbone of Google Shopping.
As you’re looking to scale you’ll absolutely need to know which queries each campaign is serving PLAs on and how this changes over time.
Get Your Key Performance Indicators (KPIs) Set
In-platform you set your bidding strategies, this may be target ROAS or max conversions.
However, these metrics are just one piece of the puzzle and a short-term view of performance.
When setting your goals for scale, think about KPIs in two groupings:
1. Short-Term Performance
Return On Ad Spend (ROAS)
Revenue ÷ Ad Spend
ROAS is often the default KPI in Shopping campaigns.
It gives a top-level snapshot of how much revenue your product ads generate for every pound spent.
But it’s only part of the story.
A strong ROAS doesn’t always mean profitable growth, especially if you’re pushing discounted or low-margin items. Brands optimising purely to ROAS often miss signals around scale and customer quality.
Cost of Sale (CoS)
Ad Spend ÷ Revenue
The inverse of ROAS, CoS is a more margin-sensitive metric that’s helpful if you work with strict profitability thresholds or marketplaces.
If your break-even CoS is 25% and you’re tracking at 32%, that’s a red flag, no matter how good the ROAS looks.
Cost Per Acquisition (CPA)
Ad Spend ÷ Number of Sales
CPA is useful for campaigns focused on driving first-time customer purchases or specific actions (e.g. buying a bundle or promotional set).
But unlike ROAS, it doesn’t factor in how much each customer spends, treating a £20 sale the same as a £200 one.
2. Mid-to-Long-Term Efficiency
Customer Acquisition Cost (CAC)
Total marketing cost /Total number of new customers
Looking beyond campaign-level spend, CAC adds in things like software, agency fees, and headcount.
It gives a true picture of how much it costs to win a new customer.
For Google Shopping, this is especially relevant when scaling into new categories or product ranges.
If your AOV is £60 and CAC is £90, your Shopping campaigns will be running a fine line on profitability unless you have strong lifetime value in place.
Marketing Efficiency Ratio (MER)
Total Revenue ÷ Total Ad Spend
MER tracks blended efficiency across all channels.
It’s helpful when you’re running Google Shopping alongside Meta, Pinterest or affiliates, especially when attribution overlaps.
A healthy MER means your Google Shopping spend is holding its own within the wider mix.
There’s no single ‘correct’ KPI, but Shopping works best when you layer short-term efficiency (ROAS, CPA) with long-term strategy (CAC, CLV) and always factor in margin, delivery, returns, order errors, etc.
Note: Our co-founder Mark wrote a full run through on KPIs, reporting and more specifically for ecommerce over at Search Engine Journal, check it out here.
Don’t Forget Dynamic Retargeting
If you’re running Google Shopping ads and not layering in dynamic retargeting, you’re leaving revenue on the table.
Shopping campaigns are great for pulling in high intent traffic, but from a new customer acquisition perspective, most visitors won’t convert on the first click (context aside).
This is where dynamic retargeting steps in to reconnect with users who viewed your products to serve them personalised ads across the Google Display Network using the item/s they added to cart, viewed, etc.
This type of retargeting is baked into Performance Max, something that is not seemingly obvious when creating new PMax campaigns, even when features such as ‘bid for new customers’ are enabled.
If you’re up and running on Shopping, structuring dynamic retargeting uses the same premise with product groupings, segmentation, bidding, etc.
Unlike non-dynamic GDN campaigns, you won’t need image/video/html creatives as it’s all powered by your product feed and the targeting such as:
- Added to cart
- Viewed products
- Previous purchasers
To get the most out of it, focus on audience segmentation and don’t treat all site visitors the same.
Someone who viewed a product once isn’t the same as someone who abandoned checkout.
Tailor your bid strategies and consider capping frequency to avoid fatigue.
You can increase spend on shopping as part of scaling but a chunk of it will slip through unless you have a plan to bring visitors back, and for ecommerce brands with large catalogues and ambitious performance goals, this tactic is low lift with high potential for return.
Fully Connected Data
Reviewing data and optimising towards a campaign or ad group level ROAS is a good start for short term performance analysis, however, it’s the data beneath the surface that holds the key to scaling.
Let’s use an example of an online clothing retailer, how many moving parts are there to consider and control that impact Shopping performance?
- Stock
- Sizing
- Colours
- Return rates
- Delivery
All of this data impacts the short and long term metrics that advertisers report on and feed into budgeting plans.
Using these factors, let’s now look at scenarios of how they feed into the day to day management of Google Shopping and why they hold such importance in being able to scale:
- Stock
- Best selling SKU limited to just 3k units one month vs 10k average
- Sizing
- Only XS, S, XL and XXL available across 30% of product selection
- Colours
- Late delivery of the black and white t shirts, the best selling colour
- Return rates
- Spiked over sale period due to 30% uptick in Klarna and Clearpay orders
- Delivery
- Company wide decision to increase next day delivery fee by 20% in month X
Each of these scenarios has a measurable impact on Shopping performance, whether it is affecting click through rates, conversion rates, or return on ad spend.
But these shifts are not going to surface when looking at platform level data, and if they do arise, it’s often too late.
Take the stock example. If your highest converting product is limited to just a few thousand units, performance may dip even though demand is still strong.
Without aligning stock levels with campaign planning, you risk spending on items that cannot scale or missing out on those that can.
With sizing, if a large portion of your catalogue is only available in less common sizes like XS or XL, conversion rates will drop.
Click through rates might stay strong, especially in visual formats like Shopping, but shoppers leave once they realise their size is not available.
The campaign appears underperforming when the issue is actually product availability.
Even small operational changes like delivery costs can have an impact and this is why fully connected data matters.
Shopping performance is not just the outcome of media decisions.
It is shaped by stock, sizing, colour drops, fulfilment, returns, and delivery settings.
Without access to that commercial context, your campaigns will always be reactive.
The difference between stable scale and wasted spend is often just visibility.
When all the pieces are joined up, optimisation becomes proactive, not just responsive.
Bringing It All Together
Scaling Google Shopping is not about throwing more budget at campaigns.
It’s about building a system where every part works together; product data, structure, bidding, measurement, and business context.
Start with the feed. If your titles are vague or your images unclear, you are already losing auctions.
Build structure around real goals. Break out best selling products, segment by margin, and use campaign priorities and negative keywords to control query flow.
Track the right metrics. ROAS and CPA give you the short term view. CAC and MER help you plan for sustainable growth. Without both, it is easy to chase the wrong outcomes.
Dynamic retargeting can help bring back high intent users and turn passive interest into actual sales.
Most of all, do not ignore the operational side. Stock levels, size availability, colour delays, or delivery changes all influence performance. When that information feeds into your strategy, you stop reacting and start predicting.
Put these foundations in place and Shopping becomes less of a guessing game and more of a reliable growth channel.
