Contact us

Your guide to shopping ads

Best practices for running effective shopping campaigns


Shopping ads are becoming increasingly popular with retailers and consumers alike. In fact, recent research by Adthena found that Google Shopping ads now drive 76.4% of retail search ad spend, in turn generating 85.3% of all clicks across text and Shopping ads.

In the UK, the statistics are even starker – with spend on Google Shopping ads amounting to 82% of overall retail search ad spend, and driving 87.9% of clicks.

First introduced back in 2009, when they were known as Product Listing Ads, Google Shopping ads allow retailers to give shoppers a stronger sense of their products than text ads, by including a photo, as well as vital information such as a title, price, store name, and more. Bing’s Product Ads act in a very similar way, giving retailers a more powerful, visually appealing presence on the search results page.

With 9 in 10 smartphone users not being certain of which brand they want to buy when they start online searches according to January 2018 research by Google / Ipsos Connect, shopping ads offer an increasingly appealing way to be top of mind when that search happens.

The pace of innovation in this field is incredible, with new ad formats being released by the engines all the time. So how can you capitalise on this and get the most out of your shopping campaigns? In this guide, we take a look at best practices for running effective shopping campaigns, to help you get ahead of the competition.

Nailing the Basics

Before you dive into the latest shopping ad formats available, it’s vital to get the basics absolutely right. Mastering the fundamentals can mean the difference between shopping ad success and failure, with many retailers missing out on some simple yet incredibly powerful opportunities.

Here we take a look at the key areas for consideration when setting up and optimising your shopping ads.

Campaign Structure

When it comes to campaign structure, it’s important to go as granular as possible to drive strong performance in the short term, whilst also keeping a focus on creating relational product groups in a tiered structure.

Segment by brand > product types

  • For example Brand > Shoes
  • Accrue data and, where the majority is coming through, break this out further, down to item ID level
  • FBreaking out hundreds of item IDs which drive 0 impressions or traffic makes campaign management difficult and does not drive performance.

Focus on best sellers and priority products

  • It is always best to focus on areas which are going to have the biggest impact on performance in the short term, with a long-term strategy also considered.
  • If your best sellers change frequently (for example every couple of months), it is advisable to avoid splitting out campaigns by product ID. Instead keep your structure at category level and label top sellers as such in the product feed, so that they pull through without having to update the campaign structure every time.
  • For evergreen best-selling products, it’s best practice to put each of these in a separate ad group, so that you can bid differently for these products, as well as enabling better traffic control and boosting click-through rate.
  • We broke out a large retailer’s top SKUs from overall products and saw a 20% increase in conversion rate (1.17% vs. 0.94%) and a 37.5% increase in ROAS (0.8 vs. 1.1)

Create a separate campaign for niche terms

  • Consider creating separate campaigns for niche terms such as:
  • Specific keywords, like dimensions
  • Terms for product brands you sell
  • Keywords for buy / purchase / order
  • Very generic terms

Consider your approach to device-specific targeting

  • Look at performance for certain keywords on different devices and if you see a trend in certain terms performing better on one device, target these terms from a different campaign excluding other devices
  • Test showing your best-selling, or highest conversion rate products on mobile devices.
  • Consider using Merchant promotions against devices where conversion rate is low.

To sum up campaign structure best practice – large inventories may mean item ID breakout is not always the best way!

  • Cover all areas at a higher level, dig deeper into core priority areas which have high investment
  • Push top sellers, high margin – scale back the majority!

An example of how we may approach this is below.

Optimising the feed

Whilst standard search campaigns and shopping campaigns differ in many ways, there is a lot to consider from the point of view of query matching. OK, you’re not matching queries to keywords, but the engines need to know that your products are relevant and match what a customer is looking for.

Getting titles, product descriptions, product type and all other fields of your feeds right is crucial to letting the engines know this. Think of each of these elements as your keywords.

This is illustrated in the table below:

Priority areas need to be your titles, descriptions and product types, as this is where you can have the greatest influence on quality score and ranking.  If your product titles seem short, test adding product categories in to see if this boost impressions. You could look to test this along with adding in specific search terms that you regularly see are triggering specific products.

Ensure product descriptions are relevant but not overly specific

  • The description should relate to a category of products in which they sit
  • Avoid using very specific descriptions
  • 500 characters is enough

When it comes to product type, use at least three nods when structuring these. For example, for a women’s knitwear retailer, this could include clothing > women > cardigans > crew neck, and so on.

Feed management software: Shoptimised

Our partners at Shoptimised, whose feed management software gives advertisers and agencies alike full control over product feed optimisation, explain this in further detail below.

Product Feed Optimisation is the process of optimising the data within your product feeds to ensure your products are shown for highly relevant search terms in channels such as Google, Bing, Facebook, Amazon, eBay and many other channels.

Optimising the keywords within your Product Titles, Product Descriptions and Product Types is the main way to influence which search terms your products will appear for other than utilising negative keywords which are only available in Google, Bing and Amazon.

Product Feed Optimisation makes up 50% of the management process for getting the best results out of your Shopping Campaigns, Marketplaces and Affiliate Networks. However, Feed Optimisation doesn’t just help to improve and control the search terms your products appear for. The way your feed is structured directly influences the way you can structure your Google Shopping campaigns which has a major impact on both growth and performance.

The quality of your product feed also has a direct impact on the performance of your shopping campaigns. The more accurate
data you provide for your products (such as genders, age groups, sizes, weights, materials etc), the better your data quality score will be in the Merchant Centre. This results in your products appearing more frequently and higher for more relevant search terms.

This is where a platform such as Shoptimised becomes invaluable, giving the ability to bulk change elements (including custom labels),  append text to titles/descriptions, test different images and run A/B testing to help drive incremental performance improvements.

Beyond the basics

Once you have your shopping campaign structure and feed optimisation in good shape, you’re already in a pretty good position.

However, there are also plenty of other tactics that you can adopt to take your shopping campaigns to the next level. Here we take a look at just some of the exciting options available, largely within Google.

Automation and bidding

If you’re running a super granular structure or have a large amount of SKUs, then automated bidding cannot be overlooked. Realistically, it’s not possible to manually optimise bids for thousands of products or product groups, so this is where the power of automation becomes invaluable. Switch on enhanced cost-per-click (CPC) and set up bidding strategies that get the machine to do the hard work for you.

Smart Shopping

Smart Shopping campaigns allow you to leverage automated bidding strategies to maximise conversions by matching a user’s intent (based on a variety of factors including demographics and search history) with the most relevant products in your feed (through remarketing or targeted new acquisition).

Below are some examples of what users would see, based on the various platforms:

Smart Shopping campaigns allow you to optimise performance based on your goals, with bids automatically adjusted to maximise conversions, as well as enabling you to reach customers on Google Search, Display, YouTube and Gmail.

Adaptive campaigns

Adaptive campaigns are a great feature in Search Ads 360 that automatically subdivides your product groups by product ID, using criteria that you define in a conversion goal or bid strategy.

This is great from the point of view of optimisation as the tech can group together products for you that drive similar revenue levels or volumes of conversions, meaning that you can adjust your bidding strategy in line with these criteria.

Merchant centre alerts

These are an invaluable feature for a shopping marketer. Setting up merchant centre alerts can help avoid five key things that will otherwise have a big impact on your campaigns;

  • Disapprovals – 5% threshold!
  • Missing required attributes
  • Long titles, descriptions
  • Missing automatic item updates
  • Policy disapprovals… these are bad.

Shopping ad formats

There are also various shopping ad formats that you can experiment with to see what drives the best results for your brand.

Showcase Shopping ads

A relatively new but powerful feature, Showcase Shopping ads give far more depth and content than a standard product ad by giving you the option to serve multiple images and expanded text with your shopping ads, showing multiple related products within a single ad, as demonstrated below.

Showcase Shopping ads are particularly useful if you are looking to:

  • Drive awareness of a new product category
  • Own a category
  • Reach new users

Trueview For Shopping Campaigns

Historically, YouTube campaigns were very much focussed on top of funnel activity and building brand awareness. This changed with the release of Trueview For Action and Trueview For Shopping.

With TrueView For Shopping campaigns, your customers can directly interact with your videos by connecting straight to your products from your video ads, bringing them closer to making a purchase.

Using your existing merchant centre feeds, up to six product cards are created that are then overlaid with your videos. The products that show can be manually controlled by selecting specific products during set-up.

Local inventory ads

Local inventory ads are designed to help shoppers find your in-store products when they’re searching near your stores. They show alongside standard shopping ads and, when clicked on, lead to a Google-hosted Local Storefront.  They can be used to promote your local inventory even if you don’t have an online shop, by using a ‘Website’ button instead of a ‘Buy online’ button.

Local inventory ads are currently available to retailers in Australia, Brazil, Canada, France, Germany, Japan, Switzerland, the UK and the US.

Spotlight on Bing shopping ads

Whilst Google claims the lion’s share of spend on shopping ads, it’s vital not to forget other platforms and engines when thinking about your shopping strategy.

With this in mind, we asked John Bodkin from Microsoft to provide his top tips on running shopping activity on Bing:

Shopping on Bing has many similarities to running on other platforms and engines, but there are three core pillars of optimisation which help to not only maximise what we’re getting from a coverage perspective, but also generate alternative user experiences which generate the best possible ROAS for an client leveraging the product:

  • Feed optimisation
  • Bid strategy
  • Reporting  

Here’s how you should be approaching each bucket:

Feed optimisation

  • Matching – Titles and descriptions take up 80% of the weight when it comes to matching the products in your feed to relevant search terms so it’s essential to prioritise these when optimising and building out exceptional Bing Shopping feeds. Other attributes which influence relevancy include GTINs, Brand, Product Category and Product Attribute.
  • Title – Remember to be specific and detailed with your product title, as you want to be sure you capture optimal intent through search terms. Example:

  • Categories – Go as deep as possible in terms of categorisation by applying as many layers of taxonomy per product as possible. For example:

  • Custom labels – Apply as many relevant custom labels per product as possible to ensure you can optimise accordingly based on priority products at the right time. For example:

  • Additional layers – We can now add a free shipping annotation, as well as Merchant Promotions into our feed to add heightened relevancy to our product ads, which should in turn increase likelihood of delivery.

Bid Strategy

We recommend that you leverage the query-level bidding strategy to ensure you’re maximising delivery and bidding accordingly depending on priority of product and likelihood of conversion.

Quite simply, the query-level bidding strategy involves the idea of separating out your products into three different campaigns by generics, brand and high-priority stock keeping units across the core pillars of Priority Settings, Negatives, and Bid Tactics.

This tactic involves the following setup per search term:

  • Generics
    • Bid = Low
    • Negatives = All brand and SKU search terms
    • Priority Setting = High
  • Brand
    • Bid = Medium
    • Negatives = SKU search terms
    • Priority Setting = Medium
  • SKU
    • Bid = High
    • Negatives = Poor Performing SKU
    • Priority = Low

This should ensure that you’re matching the right search terms to the right products, and should deliver optimal CPCs and a cumulatively robust ROAS.


We have three core reporting functions that should be leveraged to ensure that you’re maximising delivery and maintaining relevancy and eliminating wasted spend as much as possible, which are as follows:

  • Product Partition Report – This will allow you to visualise your impression share at product group level, and will give you the ability to view the benchmark bid in the auction that you can cross-compare with your own to highlight opportunity to bid up to maximise said impression share.
  • Product Search Term Report – This will allow you to optimise two core areas of performance:
    • Search terms which are generating poor performance, and thus should be added as negatives
    • Search terms which highlight particular product attributes that could be added into titles and descriptions to drive better relevancy and increased engagement
  • Product Match Count Report – This will allow you to eliminate any delivery issues in your Bing Shopping campaigns based on matching, and will help you visualise:
    • % of products submitted vs. products targeted
    • % of products submitted vs. products ready to serve

What does success look like?

Ultimately, your success metrics will be the same as they are across any of your direct response search activity, with the main goal being to drive incremental growth from your campaigns at as low a cost as possible,  by adopting a best practice approach. This can range from adopting an audience-first approach to engaging with the engines to ensure that you’re utilising all of the latest Alphas and Betas.

In the case of an audience-first approach, we were early adopters here at Croud. Looking at one of the premium fashion brands that we work with, going back to 2015 when RLSA for shopping was first rolled out, we saw immediate performance gains from heavily utilising audiences and bid modifiers. CPCs dropped by 31% and CPA by 44% within the first month!

Another feature we’ve experimented with for another premium retailer that has yielded great success, is introducing a scoring system for products, that dictates what bidding looks like based on stock availability. Let’s say, for example, that you have a product in stock that is now only available in less popular sizes. You would want these products de-prioritised for searches that don’t feature size in the query. By introducing a trade weighted availability scoring system, you can structure your campaigns around this and boost ROI. In the case of this retailer, by 17%!

New tools and functionalities are being released by the engines all the time. For example, Google recently released a new version of its Shopping Insights tool in the US, designed to help advertisers uncover more detailed information about popular brands and products. Stay up to date with the latest developments to ensure you keep ahead of the competition!



Like to know more?


Tel: 02080177723
Email: Email us
Croud Inc Ltd, Trinity,
39 Tabernacle Street, London,

New York

Tel: 13473382012
Email: Email us
Croud Incorporated,
450 Broadway, 2nd Floor
New York, NY, 10013


Tel: 61291955309
Email: Email us
Croud Australia Pty Ltd,
Belmont House, 26-28 Wentworth Avenue
Surry Hills, NSW, 2010


Tel: 02080177723
Email: Email us
Croud Inc Ltd, The Chancery,
Abbey Lawn, Shrewsbury,