How to Build Google Shopping Ads Structures that Bid to Your Business Value
As the agency that manages more shopping ads spend on Google than any other, Adlucent has always had a strong focus on driving the most from shopping campaigns. For us, this means that we take a scientific approach to maximizing value for our clients, namely testing, iterating, and re-testing product details, bids, copy, structure, and more. While best practices may change, our approach to divining those best practices has not waivered as we continuously seek the best performance for our clients. This is part three of our digital marketing automation series, which will dive into how to build structures that set your automated campaigns in Google up for success. If you haven’t already, you can check out part one here, which covers the rise of platform automation, and part two here, about balance advertiser control.
We live in a modern world of largely-automated advertising products, all designed to solve for the complexity and time it takes to manage marketing campaigns. Take Google’s Smart Shopping for example. Smart Shopping campaigns maximize conversion value based on your business goals across Google’s full reach of customers and placements, including Search, Display, YouTube, and Gmail, and combining shopping and remarketing. Machine learning optimizes to a target return on ad spend (ROAS) according to both Google’s insights and advertiser data.
The next generation of Smart Shopping is Google’s Performance Max campaigns, recently rolled out to advertisers globally. These offer a goals-based campaign experience, putting performance goals front and center using automation, machine learning, and user input. Smart Shopping and Local Campaigns will both upgrade to Performance Max campaigns next year. These serve across all Google inventory to automatically target customers where there are the best chances for scale. Ads are served at the right moments in the customer journey based on your objectives as the advertiser. According to Google, Performance Max campaigns drive +13% additional conversions at the same or lower cost per conversion when run alongside comparable campaigns. A successful example is SAIC-MG Motors, who used Performance Max campaigns to support its global expansion, specifically entering a strategic and highly competitive automotive market in Vietnam. By inputting custom segments and previous website visitors as audience signals, the brand worked with Google to help speed up automation’s ability to optimize results while allowing Performance Max to find new customers. This led to a 39% increase in test drive leads at an 83% lower cost-per-lead compared to the account average. Google Performance Max campaigns are now part of SAIC-MG’s core media strategy.
Recently we published a Commerce Digital Marketing Maturity blog post,where we discussed how important automation is in gaining a competitive advantage and outperforming, but what we are going to dive more into here is how crucial it is to properly structure your campaigns, product data, and content to drive the best results.
In managing performance advertising for over a hundred large retailers and brands at Adlucent, we have found that a simple structure that lumps all products and goals together doesn’t provide Google with as much guidance for understanding your ideal shoppers as a structure that aligns the product connections to those shoppers can provide. In short, we find a Smart Shopping structure that accounts for various customer types and business goals will typically allow the platform automation tools to understand more information about what you’re trying to accomplish from the program, and ultimately drive more business value.
While the “levers” of control are much more limited for these campaign types than they have been for more traditional shopping campaigns, there are four key areas of control that we leverage as advertisers to drive business value for our clients from these programs::
Campaign Structure: How do we group products together in campaigns?
Budget: What is the budget for each campaign?
Goal: What is your target KPI each campaign should be driving to?
Conversion Value: What is the value of each conversion you receive from the campaign?
Below are recommendations for improving campaign structures to help to drive more success from these campaign types.
Structure For Success
In the testing we have done at Adlucent, helping to create product associations by grouping them together in campaigns has helped Smart Shopping automation find more of the right customers versus simply lumping all the products all together in a single campaign. This was proven on our client, Gardener’s Supply Company when we leveraged Smart Shopping campaigns to break the performance plateau typically experienced in their traditionally slower winter season. We grouped together products in Gardener’s holiday gift collection, then separated them into their own Smart Shopping campaign in order to find the most relevant customers possible. The result was a 756% profit increase and a 465% increase in revenue from their holiday gift collection year over year.
The number of campaigns that you separate across a given program is largely dependent on what type of associations work best for your products and the overall scale of the program. Generally speaking, large-scale and enterprise-scale accounts can greatly benefit from adding more granularity, so long that there is still enough spend and traffic going through any given campaign to ensure that there is still enough data feeding the automation.
The idea here is to consider looking for some of the highest success categories or product groups so you can manage budgets and goals at a more granular level. A few examples of how advertisers might want to consider looking to segment products for Smart Shopping account structure:
Top Sellers: Campaigns for the top-selling categories or product groups can provide more control over some of your biggest spend and return areas. Looking simply at sell-through or revenue-per-click is probably more important than looking at efficiency (eg: ROAS) of categories since the goal is to identify high-volume categories.
High Return Products: Consider testing putting your products with the highest rate of return in a distinct campaign so you can better control budgets and goals associated with products that may be more costly for your business.
Seasonality or Geography: If there are multiple distinct selling seasons across your product set or if selling varies greatly by geography throughout the year, these two views may be a more important way to segment your campaigns.
Pro Tip: If you’ve not yet transitioned to Smart Shopping or Performance Max, looking at query data volume can help identify how to group products based on the types of queries that had surfaced particular SKU sets.
As we’ve mentioned in our previous post in this automation series, current platform automation doesn’t handle multiple simultaneous goals very well. So, if you’re an advertiser that is both selling online and driving store traffic, platform automation may have difficulty rationalizing two distinct goals as smoothly as it can handle focusing on just ecommerce sales or just in-store traffic/revenue.
Adlucent’s recommended solution is to separate Local Inventory Ads (LIAs) from your Google Smart Shopping campaign in order to drive in-store metrics from a separate campaign with goals focused on store traffic, visits, or in-store revenue.
Google provides flexibility to allow LIAs to be excluded from Smart Shopping Campaigns so that they can be managed independently via a separate campaign.
With Google’s recent announcement around Performance Max Campaigns, they do indicate that Local Campaigns will be integrated into Performance Max, so understanding the impacts of managing Performance Max campaigns to both ecommerce and store sales is something that will require more testing to understand the ideal setup for maximizing both in-store and online sales. However, if you are an advertiser that requires distinct reporting of costs against in-store vs online transactions, distinct campaigns will still need to be used to distinguish reporting of spend between the two.
Embrace a Structure (And Culture) of Testing
Smart Shopping campaigns (and Google’s more recently unveiled Performance Max campaigns) are still rather new in the overall landscape of Google Shopping solutions. We know that the automation is geared to help us find more customers, but understanding exactly how to better inform that automation is still something that each advertiser should look to gain for their business.
As you consider your structure, including an approach that allows for an always-on geo split test of some sort will let you to test budget and goal changes and injections of spend in a specific portion of the program before it rolls out to the entire account.
We’re wishing you a Merry (Smart) Shopping Season!
Keeping these recommendations in mind will help you better leverage Google’s high-performing, quickly growing Smart Shopping solutions. It’s important to remember that optimizing Smart Shopping or Performance Max structures can be an iterative process, starting with one category or product grouping at a time.
While Google is currently leading the industry in this area, automation algorithms in ad platforms will continue to evolve and get better at identifying the right customers for your business – testing for the best performing product structures can potentially be leveraged for other platforms, as well.
Our team continues to engage in ongoing testing around Smart Shopping, Performance Max, and automation. We plan to continue to share our findings and results of our tests in the coming months. If you have questions about the results we’ve seen in building more granular Smart Shopping and Performance Max campaigns, don’t hesitate to contact us to talk about the right approach for your business.
Ryan is VP of Strategy at Adlucent. He has over 15 years experience working with digital and multi-channel marketers to drive better results from their programs. His perspectives of online marketing are based on analyzing data-driven insights from an array of advertisers - from enterprise to start ups - as they capitalized on customer-centric, full-funnel programs across search, social, shopping and digital media.