Should You Jump Aboard the Smart Shopping Hype Train?

In recent times, Google’s marketing machine has gone into overdrive to convert old school traditional Google Ads users to the magic of smart bidding. This has permeated everything, from the optimisation of individual keyword bids to the selection of the highest converting ad headlines and descriptions. Despite the initial scepticism from the digital marketing community and reluctance to hand over the reins to the machines, the transition has proved to be a resounding success. Within our agency, we have witnessed first-hand the incremental gains achieved by smart bidding strategies across a range of verticals. With the algorithms doing most of the heavy lifting, this allows analysts to devote more attention to wider digital strategy and the challenge of scaling accounts.

What exactly is Smart Shopping?

Some of you may be thinking, “Wait a minute, don’t we already have target ROAS smart bidding for Shopping campaigns?”
Smart Shopping is a more evolved animal. Not only does it incorporate smart bidding, it also introduces a Google Display network component in the form of dynamic prospecting and remarketing. Smart Shopping campaigns extract product information from your Merchant centre feed and combine it with your uploaded image assets to create various ad types. Machine learning will then figure out the best ad placement, time and audience to target based on your business goals. These ads can show across Google Search, Display, YouTube and even Gmail.

Getting Started

Smart Shopping campaigns are surprisingly simple to set up. All you need is an active merchant centre product feed, a logo image, a landscape image and text. The text is made up of a short headline (25 characters), long headline (90 characters) and description (90 characters). Your account will also need at least one remarketing list containing a minimum of 100 users.
Detailed step-by-step instructions on how to create a Smart Shopping campaign can be found here.
Once your campaign is live, you should aim to let it run for a minimum of two weeks. The more data you have, the more reliably you can assess its performance.


Here’s how Smart Shopping is driving performance for one of our ecommerce clients. The test was conducted on our Bestsellers products campaign and went live in September 2019.

Previous Period (Jul - Aug 2019 compared to Sept - Oct 2019)
YoY (Sept 2019 - Oct 2019 compared to Sept 2018 - Oct 2018)

In both cases, Smart Shopping clearly led to significant increases in ROI, 13% for the previous period and 72% YoY. There were also notable reductions in CPC and CPA. Now that we’ve seen what Smart Shopping can do, let’s take a look at some pros and cons.


  • Machine learning – the most obvious benefit is that machine learning optimises bids in real-time for each unique ad auction. A huge number of signals are weighed as inputs making it far superior to manual bid adjustments made by humans.
  • Time saving – less time managing bids means more time for other tasks.
  • Built-in dynamic remarketing – this simplifies advertising across Google’s network for less experienced advertisers and allows them to effectively draw customers back to their online store.
  • More reach – Smart Shopping is not limited to search demand. Showing ads on the Display network and YouTube allows advertisers to reach a much large audience. Perfect for scaling up.
  • Hybrid bid strategy – this is an interesting one. If you look in the campaign’s bid settings, the smart bidding strategy runs on maximise conversion value but also gives you the option of setting a ROAS target. Is this a glimpse into the future of smart bidding?


  • Less control – with the machine changing the bids, sometimes it can be difficult to drive more spend to a specific product you want to promote. Especially, if the product doesn’t perform as well as others in its product group in terms of ROAS or raw revenue.
  • Less granularity – one solution to the above is to split out your products into as many campaigns as possible so you can set different ROAS targets for each. The drawback here is that Smart Shopping, and all smart bidding strategies in general, work best when they have lots of data to work with. Accounts with smaller budgets spread across multiple campaigns will generate less conversion data for the machine to learn from. There is also a maximum limit of 100 Smart Shopping campaigns per account.

In conclusion, yes, you should jump aboard the Smart Shopping hype train, or at least ride it for long enough where you can evaluate its performance. For the majority of advertisers, surpassing ROAS targets is the ultimate measure of success and Smart Shopping has consistently proven its value here. Furthermore, smart bidding can only get better as Google continues to innovate and push new updates every year. While there are some drawbacks for advertisers who enjoy complete control, this will likely be a minor thorn in their side when compared to the vast benefits. Let us know your experience with Smart Shopping in the comments below!