This website uses cookies to ensure you get the best possible experience. See our Cookies Policy.

PMG Digital Made for Humans

PLA-ying Around With PLA Structures

6 MINUTE READ | July 19, 2017

PLA-ying Around With PLA Structures

Google Shopping is becoming an ever more important piece of the Google Search landscape. From changes in SERPs to better highlight shopping to larger percentages of budgets shifting to shopping, PLAs aren’t going away anytime soon. In fact, according to industry reports, traffic and investment KPIs are trending upwards for PLAs. Shopping ads now account for over half of all Google Search clicks and 75% of all non-brand searches. With this increased importance, an intelligently structured PLA account is a must. This doesn’t mean there is only one correct way to structure an account; rather there are a multitude of options. The type of client, investment level, and overarching search strategy all affect this. Below we will look at the retail space and a few structures to consider when setting up a new PLA account or looking to enhance an existing one.

Not everything in advertising needs to be complicated – there’s merit in simplicity. Product offerings for a retail client naturally fall into a number of broad categories that will remain consistent for most accounts. The most efficient way forward is to rely on these established categories. We typically assess our client’s offerings and come up with 10 – 20 broad categories we can use as campaigns. For example, dresses, shirts, jeans, and pants.



We structure our product feed and campaigns based on these broad categories and then create one ad group within each campaign that mirrors the campaign name. While multiple ad groups can be useful, we have so much control at the product group level that multiple ad groups aren’t necessary. With broad categories at the campaign and ad group level, we start breaking out specifics at the product group level. For example, s__kinny jeans, high-rise jeans, boot-cut jeans.

By keeping the broad categories at the ad group level, all settings and modifiers will remain consistent by category. This makes management and reporting simpler which allows for quicker changes across the account. When we need or want to add another level of specificity while still keeping things broad, we typically split our campaigns by gender to provide better product breakdowns and insights.

Structure 1 is a great option, but it can sometimes lack the level of granularity or specificity we need or want in our PLA structure. In those cases, we make modifications to Structure 1 that allow us to be more specific, while still maintaining the overall principles laid out in Structure 1. So while Structure 1 and Structure 1B are similar in the basic principles of their set up, there are some key differences.

While multiple ad groups and specificity are not necessary, that doesn’t mean they can’t be beneficial. A good way to add some granularity without being overwhelming is to remove some specificity at the campaign level and move more to the ad groups. Using this structure, instead of structuring our campaigns with product categories, we use even broader groupings such as tops, bottoms, shoes, and accessories. This allows us more control at the ad group level without creating an excessive number of campaigns.

With these broader campaigns, our ad groups will look different. Now, ad groups will focus on the subcategory level. For example, in bottoms, we might see jeans, joggers, chinos, and sweatpants. This includes everything that previously would have been in jeans and pants but still moves some specificity up a level from product group to ad group. Feature specific breakdowns will still live at the product group level.

The benefit of this structure is that more control is given to more specific groupings. We can now manage bids and settings specifically for chinos and joggers rather than just pants as a whole. This allows us to affect device modifiers, geographic modifiers, audiences, keyword negatives, and more for a more granular set of products. For some clients or product types, this may be unnecessary, but we have found in some cases that certain sub categories perform differently enough to warrant using different modifiers and settings at a more specific level.

Another trend that is becoming more and more popular is focusing on the audience the ad is serving to. With this in mind, we can structure our account to first concentrate on the user’s we’re serving to, and the product categories second. Users who have never visited the site, users who have visited but not purchased, and users who have purchased all behave very differently and deserve different treatment.

Instead of breaking out our campaigns based on product categories, in this structure, we break them out based on the audiences that are most important to our client’s businesses and apply those audiences to each campaign selectively. This will vary from client to client, but some common audiences to include are: new customers, site visitors, cart abandoners, past purchasers, and lapsed purchasers. We can now tailor everything about the campaign towards the users in each group.

For example, maybe we only want to show new customers a particular subset of the product catalog through inventory filtering or want to limit budget spent on site visitors that never made it to the cart, both of these things can be achieved at the campaign level. You may find that the difference in performance between different audiences is much more noticeable and actionable than the difference between product categories.



With all that said, we still need to target different product categories at the ad group level, so we don’t bid the same amount on the entire catalog. The ad groups can mirror the setup from option 1 or 1B, but if we’ve used the more granular/specific option (1B), this will create a large number of ad groups per campaign. The main benefit of the “thinking outside of the box” lies in being able to customize all parts of the campaign and experience for each type of user rather than just bidding up or down on audiences at the campaign or ad group level.

Insights meet inbox

Sign up for weekly articles & resources.

These are just three of many different account structures that could work for any given client. Even if you don’t use one of these strategies specifically, the overarching principles apply to any search account. The one thing to always keep in mind is we never want to make our client’s accounts unnecessarily complicated. Our goal is always to provide the level of granularity that both the client and the PMG team is comfortable with, and the level of specificity that best meets our client’s needs.


Posted by Garrett Milliken

Related Content

Get Informed

Google Simplifies Smart Bid Strategies

2 MINUTES READ | April 23, 2021

Get Informed

Google Paves the Road Ahead for Responsive Search Ads

3 MINUTES READ | February 18, 2021

thumbnail image

Get Informed

Google Phases Out Broad Match Modifiers

4 MINUTES READ | February 4, 2021

Get Informed

The Road to Recovery for the Travel Industry

5 MINUTES READ | November 19, 2020

Get Informed

Price Glomski Talks to Hospitality Technology

1 MINUTE READ | October 28, 2020

thumbnail image

Get Informed

The United States of America v. Google & What Comes Next

5 MINUTES READ | October 22, 2020

Get Informed

A Closer Look at the Congressional Big Tech Market Power Report

6 MINUTES READ | October 19, 2020

Get Informed

Auction Time versus Intraday tROAS Bidding Strategies

4 MINUTES READ | October 5, 2020

All POST