Debate: Do Single Product PLA Ad Groups Strangle ROI?

Google ShoppingBuilding out Product Listing Ad campaigns in Google AdWords with single product or single SKU ad groups has become a fairly common practice. The idea being that isolating each product into its own ad group gives advertisers more control over which item gets served for a given query and enables faster product-level performance analysis and more granular bidding strategies.

A new paper from CPC Strategy challenges this approach to PLA management. It’s not a new position from CPC Strategy, but CEO Rick Backus says,

“We were surprised to see how quickly this SKU-level build out strategy gained traction in our industry. Internally, our tests showed that it was a very basic and primitive way to structure PLA campaigns, but somehow it’s perceived as this ‘cutting edge’ strategy. We pretty much set out to debunk that perception.”

The company says the single product PLA ad group approach is:

  • Difficult to manage effectively even with bid automation in place when there are potentially thousands of separate ad groups
  • Generally don’t maximize impression share for products
  • Spread ad spend too thinly without taking into account specific conversion rates among similar or top-performing products.
  • Limit opportunities for increased product exposure by isolating high quality score across individual products instead of grouping them within related product groups

Case Study

After testing single product ad groups across several clients, CPC Strategy found that in most cases the approach hurt ROI. In one case study, CPC Strategy restructured the PLA campaign for their client, Clean It Supply, into single SKU ad groups to compare before and after performance. Compared to the previous period, in the three week after implementing the single SKU structure, in conjunction with campaign automation:

  • Traffic fell 34 percent
  • Revenue was off 57 percent
  • Average order value was down 58 percent
  • Conversion rates dropped 36 percent

Performance rebounded after reverting back to CPC Strategy’s standard PLA management.

Advocating Mixed Strategy

CPC Strategy does recommend creating single product ad groups for top performing products when an item stands out among a group of related products. The advertiser should pick that item out of the ad group and place it in it’s own ad group at a higher bid and monitor it manually. Yet, even then, CPC Strategy contends that the retailer should be prepared to add it back into the theme-based ad group if performance doesn’t improve with separation.

The company recommends creating a Top Performers ad group, typically limited to the top 10 percent of inventory to maximize volume and ROI. That 10 percent number might cause raised eyebrows — that’s potentially a very large number of products in an ad group.

Yet, based on their studies, CPC Strategy believes that maintaining themed multi-product ad groups creates a boat lifting effect: 30 products will perform better  in a  Top Sellers ad group than in single SKU ad groups.

The company then typically groups the remaining inventory by product type and brand — or by other groupings that make sense for the business, such as price range or season by using custom labels. Poor performing products are kept in the feed in their own All Product Target ad group at a penny bid. Essentially, they’re advocating a mixed strategy approach.

There are many advocates for single SKU ad groups in PLA campaigns — and for reasons beyond what is outlined above.  Which approach have you seen work?

Related Topics: Channel: Retail | Google: AdWords | Google: AdWords: Product Listing Ads | Search Engines: Shopping Search Engines | Search Marketing: Shopping Search Marketing | Top News


About The Author: writes about paid online marketing topics including paid search, paid social, display and retargeting. Beyond Search Engine Land, Ginny provides search marketing and demand generation advice for ecommerce companies. She can be found on Twitter as @ginnymarvin.

Connect with the author via: Email | Twitter


Get all the top search stories emailed daily!  


Other ways to share:

Read before commenting! We welcome constructive comments and allow any that meet our common sense criteria. This means being respectful and polite to others. It means providing helpful information that contributes to a story or discussion. It means leaving links only that substantially add further to a discussion. Comments using foul language, being disrespectful to others or otherwise violating what we believe are common sense standards of discussion will be deleted. Comments may also be removed if they are posted from anonymous accounts. You can read more about our comments policy here.
  • MerryWhy

    Hey Ginny,

    The link is broken here, but great stuff!

  • Ginny Marvin

    Fixed! Sorry about that.

  • MerryWhy

    Thanks, you are awesome! Excited to hear what strategies people are using on PLAs!

  • Sam Mazaheri

    Ya, I’d love to hear everyone’s opinion on this debate

  • Doug

    As a product photography strategist I do find value in A/B testing the images of identical products using specific groups, but generally once a particular style of photo is working I group the number of related products accordingly.

  • George Michie

    We manage more PLA spend than any other agency. We find that there is no one-size-fits-all approach that maximizes performance. Product level ad groups work best for some advertisers. Some benefit from more aggregated groupings. The mistake people make is “set it and forget it.” You have to test different targeting strategies to find what works best for a particular account.

  • Mark Roudy

    I am already running my campaign in the same path….Thanks

  • d_a_t

    Were the old targets simply paused when they launched the new Prod ID targets? In this case, you would expect to see a drop off in performance as the new targets have no history and thus less quality in Google’s eyes. By reactivating the old targets, which is what I assume happened in the recovery phase, you get all of that history back. Thus, this could just be a case of the older targets performing better because they’re older, not because using Prod IDs doesn’t work.

  • Hamid Saify

    It’s a similar approach that was used for keywords in the mid 2000s. One keyword per ad group, match type, most relevant creative. Of course, that become a nightmare to manage & with matching updates, unnecessary. But, the proof here is what? One case study, for one client, in a highly specific vertical. I’d hardly call that case study material refuting a strategy. But well played, I read it.

  • Josh Brisco

    It is the product IDs, GTIN and general data feed information which carry the product relevancy and not the targets. In taking over many existing campaigns and building out a new one from scratch with new product targets, I can say that there is usually little to no negative change in performance with the new targets in place.

  • Rick Backus

    George – I 100% agree and this is exactly what we say in the guide (excerpt below):

    “As with all online advertising there is not one strategy that guarantees optimal results in every case. smart retailers looking to get the best return on their ad spend understand the importance of testing and evaluating multiple strategies.”

    The point of the guide was to debunk the myth that SKU level build outs are the most sophisticated way to manage a PLA campaign. It’s all about getting the client the best ROI possible and if the ONLY strategy is a SKU level build out, you’re doing it wrong.

  • Rick Backus

    @hamidsaify:disqus that’s a fair critique and I could see how you could feel duped. The data in the guide is only for one of our clients but we manage over 30 million a year in ad spend across 200+ clients and we have seen this trend across the majority of our clients. We should have made that evidence more clear in the guide.

  • d_a_t

    Product relevancy does lie in the data feed, but quality score does not. Having seen specific instances of product IDs being launched at significantly higher bids than broader targets with more history, yet the broader target continues to get all of the traffic for that product (we database the final destination of every click to match up what products are served through broader targets), indicates to me that Google does assign some sort of quality score to the target itself based on history. Otherwise, why would they serve the broader target with the lower bid?

    Also, your study’s conclusion is that launching all of these product IDs did hurt performance, which seems in contrast with what you’ve just said, ‘that there is usually little to no negative change in performance with the new targets in place.’

  • Lucas von Fürstenberg

    With an ever changing product inventory I see value in grouping SKUs by category and brand. Especially when going as granular as having separate targets for every size available, grouping them together ensures that Google at least has some idea as to how a target might perform. Otherwise you may end up with very little data per adgroup.

Get Our News, Everywhere!

Daily Email:

Follow Search Engine Land on Twitter @sengineland Like Search Engine Land on Facebook Follow Search Engine Land on Google+ Get the Search Engine Land Feed Connect with Search Engine Land on LinkedIn Check out our Tumblr! See us on Pinterest


Click to watch SMX conference video

Join us at one of our SMX or MarTech events:

United States


Australia & China

Learn more about: SMX | MarTech

Free Daily Search News Recap!

SearchCap is a once-per-day newsletter update - sign up below and get the news delivered to you!



Search Engine Land Periodic Table of SEO Success Factors

Get Your Copy
Read The Full SEO Guide