Get Your Priorities Straight: Structuring Google Shopping Campaigns
A good campaign structure can have a significant impact on performance, but what structure is right for your business? Columnist Laura Collins discusses some options that have worked well for her clients.
I’m a big fan of Google Shopping. As a user, it provides an easy-to-use platform for comparing products and prices at a glance. Instead of clicking through to eight different websites to find the item I want at the right price, I could make my choice before I even leave the search results page.
Shopping campaigns are invaluable for any e-commerce retailer. At my employer (Periscopix, a Merkle Company), we consistently see them outperform generic search in terms of revenue and cost of sales.
Shopping campaigns have taken many shapes over the years, as Google has responded to demand and made improvements. Many of you may remember Product Listing Ad (PLA) campaigns, Shopping’s previous incarnation.
While creating a campaign structure based on your feed was possible, it certainly wasn’t as simple and intuitive as it became when Shopping campaigns were launched in spring 2014. Also, a brand-new feature was the ability to use campaign priority settings in order to preferentially serve particular products over others, without the need to increase bids.
While at first, the more complex options at your disposal with Shopping seemed daunting in comparison to relatively simple PLAs, we soon found that they enabled us to create campaigns that reflected our clients’ needs, were easier to optimize and performed significantly better.
There are a variety of ways you could choose to structure your Shopping campaigns, but how do you know which one is right for you?
With our diverse range of retail clients, we’ve had the chance to try out dozens of different structures and analyze performance. Here, I’ve collected some of the best and hope to provide inspiration for anyone who’s struggling with deciding which structure is right for them.
One of the most exciting new possibilities with Shopping campaigns was the ability to assign different priority levels to campaigns, in order to favor certain products over others. We’ve explored several ways of prioritizing our campaigns to achieve the best results. There are a few that we’ve found to be highly successful.
This is a structure based on the need to (a) promote sale products over those offered at full price; and (b) to report on generics and brand separately. Adding a “Sale” custom label to any products on offer within the feed allows us to add those products to a high-priority Shopping campaign.
This means if someone searches with a term that could apply to two products, one on sale and one at full price, the sale item will be favored. We know a discounted item is more likely to encourage someone to buy, so we’ve found this helps boost conversion rates.
The medium- and low-priority campaigns are then divided by generics and brand respectively. Adding all brand terms as campaign-level negative keywords to a medium-priority campaign filters brand traffic down to low priority. This gives the ability to optimize towards different KPIs for brand and generics, and it makes reporting more transparent.
2. Promotions/Best Performers/All Products
This structure has proved useful when pushing best-performing products is a top priority. The high-priority campaign is the same as above, created using custom labels applied to sale or promotional items.
Then the medium-priority campaign is built by splitting products with the strongest performance down to item ID level and leaving everything else to filter into the low-priority “all products” campaign.
Regular analysis of the low-priority campaign helps identify items that are performing well, which can then be moved up into the Best Performers, ensuring products to which we know users respond well are appearing in the Shopping results most.
3. Long Tail/Short Tail/Brand
This structure was thought up as a way of increasing impression share on generic terms without wasting spend on users too far up the purchase funnel. The high-priority campaign was designed to capture people who’d already done their research and knew exactly which product they wanted to buy (e.g., someone searching for “size 8 red striped maxi dress” rather than “dress”).
This was achieved by excluding short-tail search terms and maintained by regularly checking search query reports to ensure just long-tail search terms were being captured.
Short-tail queries then filtered down into the medium-priority campaign to capture users who were still in their research phase. Brand terms were added as negative keywords to both the high- and medium-priority campaigns to allow brand traffic to trickle down into low priority.
By segmenting search queries in this way, the team was able to prioritise long-tail search queries and increase generic impression share specifically for those terms, where they knew they were more likely to lead to a conversion.
Once you’ve decided how to use campaign prioritization, you need to think about what structure you’re going to use within each campaign.
Ad Group Or Product Group?
Whereas in a traditional search campaign, you would build your structure on ad groups and keywords, in Shopping, you’ll find ad groups and product groups. However, that doesn’t mean product groups are the same as keywords.
Product groups let you carve your feed into smaller segments to allow for more granular optimization. If you didn’t add any keywords to a campaign, you’d be targeting nothing, whereas your Shopping campaign would target everything within your feed without product groups in place. They allow you to build a logical structure that is easier to manage and optimise and to exclude any products you don’t want included.
While you can set bids at both product group and ad group level, there are several instances where you may need to use ad groups. Mobile bid adjustments, promotional text and negative keywords are all applied at this level, so it’s a good idea to group different product ranges into different ad groups to allow for relevant promotional text and easy exclusion of irrelevant traffic.
Subdivide All Products By… What?
Once you’ve created your first ad group within a Shopping campaign, you’re given the choice to subdivide “All products” into separate product groups, which you certainly should do. While an “All products” ad group is useful as a catch-all tool to ensure complete Shopping coverage, relying on it entirely leaves you with no ability to analyze performance and optimize.
So which option do you choose for subdividing your products?
- Category. This is probably our most commonly used option. It allows you to divide your campaigns by the Product Category attribute within your feed. The picture below is an example of what a typical product category structure could look like for an online clothes retailer. So in order to create a product group for maxi dresses, you would need to subdivide into Womenswear > Dresses > Maxi Dresses. This relies on a healthy feed with logical Product Categories in place, so if this type of subdivision isn’t working for you, it could be time to make some improvements to your feed.
- Brand. This is particularly useful for retailers that sell multiple brands and want to bid or report on them separately.
- Item ID. This is the most granular you can get. Unless you have a very small selection of products, subdividing to item ID level certainly isn’t necessary for all of them. However, it can be useful once a campaign has been running long enough to gather data, if you notice particularly strong or poor performance from specific items.
- Condition. Condition is relevant for retailers selling both new and second-hand/used products.
- Product type. This allows for more granular categorization of products within your feed, if the options within the Product Category attribute aren’t specific enough for your needs. With a similar hierarchical structure, you can subdivide by Product Type in the same way as with Category.
- Custom labels. These labels can be applied to products within the feed in order to separate them within Shopping when it isn’t possible using any of the options above.
Finding the campaign, ad group and product group structure that works best for you should provide you with highly granular Shopping campaigns which are easy to manage, analyze and improve.
There is no “right” way to structure your Shopping campaigns. It depends entirely on your business’s reporting needs and KPIs.
The great thing is that there are so many options at your disposal now that with a little consideration, you should be able to find the perfect one.
Opinions expressed in this article are those of the guest author and not necessarily Search Engine Land. Staff authors are listed here.