Today’s post is by Joey Muller of CPC Search.

After nearly six months of declining Google Shopping-related traffic, changes to the Google Shopping model, and an industry-wide scramble to recoup lost revenue, ecommerce advertisers are showing signs of resilience. There is evidence of lower CPCs, higher profit margins, and rising spirits across marketing departments that not long ago scrambled to answer the question, “Where did our Google Shopping revenue go?”

October 1st marked the completion of a six-month transition from free listings to Google’s pay-to-play solution known as Product Listing Ads (PLAs). While many advertisers have already transitioned, some folks are just learning about PLAs for the first time. Let’s do a quick recap of how we got here and then look at some best practices for mastering this new solution.

What are PLAs?                                                        

PLAs appear in the sponsored results of and require advanced bidding and advanced data management expertise. They consist of an image, title, description, price, and URL, and serve a single purpose: to better connect consumers with retailers and products. Like traditional text ads, we manage bids and campaign structure using AdWords; unlike text ads, we manage ad copy inside a product feed. For example, to change an item’s price, we change the product feed.

PLA example

Examples of PLAs

While not new, PLAs have come to define the new Google Shopping experience. That’s right – the old Google Shopping experience is now defunct!

In the old Google Shopping, retailers enjoyed free traffic from Google’s organic products listings, in some cases bringing in 30% or more of total online revenue. This was a sweet deal for retailers. However, with the new model, there is no free lunch. Gone are the days of easy organic product listings; they’ve been replaced by a complicated, multiple-touch system where success depends on smart bidding, proper account structure, and a healthy product feed.

To master PLAs, you must first master Product Feeds.

Filling in your feed with data requires time and forethought. There are required fields: id, title, image, description, url, and GTIN. There are recommended fields: Google product category, product type, additional image link, sale price, and sale effective date. And there are optional fields: brand, mpn (manufacturer part number), gender, age, AdWords grouping, and AdWords labels.

The more fields you maintain, the more bidding control you will have in AdWords, but the more likely you will encounter failed uploads. (Every time you make a change to the feed, you must re-upload it.) Just be sure to check the Status area of the feeds tab of your Google Merchant Account to see how your last upload fared (see screenshot below).

PLA upload feeds

Google recommends that to keep things fresh, we re-upload our feeds every one or two weeks. Between choosing the right fields, monitoring status messages, and uploading often, it is practically a full-time job just to keep product feeds healthy. If you’re a large retailer, you know just what I mean!

Set up AdWords for PLAs.

On the AdWords side, there are four steps to get started:

1. Create a dedicated PLAs campaign. Do not mix PLAs with other search campaigns, else tracking will get very confusing very quickly. (Decide on an Ad Group structure that makes sense for you—making this decision is probably worth another blog post all by itself, but let’s skip it for now.)

2. Create a new Product Extension on the Ad Extensions tab. This is where we tell AdWords which Google Merchant Account to reference for our product feed. If you have already specified the AdWords customer ID for your AdWords account in your Google Merchant Center account, you will see an available merchant account id in AdWords.

3. Create a new Product Listing Ad from the Ads tab. This goes into each ad group in your PLAs campaign, and you can copy and paste this for all ad groups in AdWords Editor. Promotional text is optional, but if you have free shipping or a 100% guarantee, do mention it here.

4. Add product targets in the Auto targets tab and set a bid for each target (more on this below).

Focus on Product Targets.

First, create an All Products target. This is a catch-all that enables full coverage for every product in your feed. Since you don’t have any control over what Google will show, you may want to set a relatively low initial bid.

Next, think about your top-performing products and create a target for each one. You can do this at the brand level (brand=Sony), at the brand + product category level (brand=Sony, product_category=Laptops), or at the individual SKU level, where product ID matches a single SKU (id=SVE1113FXW). Bid differently on these targets according to your economics (average order size, gross margin, target ROI, etc.). Note that for any bids that are lower than your catch-all, all products bid will probably be superseded by your all products bid. Also note that at the end of the day, even if you are using product-id-level product targets, Google still controls which user queries are mapped to which PLAs. Don’t think that using product-id level targets is similar to using exact-match keywords; it is not the same.

Each time you set a target, be sure to click Validate. If you enter values exactly as they appear in the feed—punctuation included—your targets will validate.

Tip: The most recent version of AdWords Editor now supports targeting individual product IDs. This means you can take your entire feed and upload it into AdWords as follows:

1. Open your feed in Excel.

2. Remove all columns except id.

3. Add a new column called Max CPC and enter bids for each id according to your economics.

4. Copy your columns and paste them into AdWords Editor. (There are additional nuances here—it took us several tries to get it right—but once you nail it, the ability to bid at the SKU level is a very powerful thing.)

Where do I put my keywords in AdWords?

True or false: keywords play no role in PLA campaigns. True! It’s hard to fathom, but in this case we just leave it to Google to match queries with products. This illustrates the importance of optimizing your product feed. Take time to put important keywords in your titles and at the beginning of descriptions, and remember Google will bold-face matching search terms in your PLA.

Negative keywords, on the other hand, do play a role in PLAs. Run search term reports to see what queries are triggering your ads and use this data to build a solid negative keyword list (see screenshot below).

adwords search term report

Put it all to good use.

That’s it – now you can work on bid optimization, as usual. Good luck!


Joey Muller CPC SearchJoey Muller is a certified Google AdWords Professional at CPC Search, a PPC agency in San Francisco. Joey has generated millions of dollars in revenue for ecommerce, gaming, healthcare, and IT companies. Joey received his BA in Cognitive Psychology from Dartmouth College. See Joey’s LinkedIn profile and follow him on Twitter @jmthefourth.

This entry was posted in Google. Bookmark the permalink.

9 thoughts on “Master Product Listing Ads (PLAs)

  1. SKU level bidding sounds amazing! Did you see better returns when you switched to SKU level bidding? Do you still keep an ad group that targets all products at a low bid?

  2. Amanda, we have seen very good results with SKU level bidding, especially where we already optimize for profit. For ex, SKU123 is a product very similar to SKU456, but 123 has twice the profit margin so we bid accordingly.

    We always keep a single ad group assigned to All Products and bid it low. If you took all of your SKUs and brought them into AdWords Editor, you could argue there is no longer a need for All Products. But just in case, we like to keep All Products around.

  3. Any chance you can get more specific on how to upload the skus to AdWords editor, or how to build out a sku driven product ads campaign step by step? That would be awesome!

  4. I saw a quick preview of Adlucent’s Product Listing Ad Management solution at eTail this week and was extremely impressed with the solution. I really liked the use of consumer intent analytic underpinning that can drive significantly better PLA performance.

    Has anyone seen this offering, or any others that looked promising from a PLA management perspective?

    Thanks, Bruce.

  5. Hi Joey

    Wow, at last. An article that clearly sets out how to combine product attributes. Many Thanks.

    The paid Shopping Model is now being phased in in the UK. I have had a PLA Campaign for a year now using only Brand and Product Type, separately, as Targets. It is now time to get more granular control on bidding.

    We have over 15,000 products in our feed. Am worried that this number of targets will be unmanageable. How would you suggest organizing such a Campaign using Product id?



  6. Hi Joey,

    Really enjoyed the post. PLA revenue is growing at an exponential rate and this post does a great job highlighting ways to reach PLA goals.

    One question though. The adwords_label attribute in the data feed accepts up to 10 labels separated by a comma. Does that mean that I can target a single label?

    I haven’t had the opportunity to test my feed yet but I would love to structure my campaign around these labels. I plan on including labels specific to seasonality, high-margin/low-margin, and product details to name a few. I think this will allow me to test and optimize different combinations for the best results.

Leave a Reply

Your email address will not be published.


You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>