Yesterday at NetElixir, experience and curiosity were the toppings for the fresh hot evening coffee for four enthusiastic search marketers. It all started during a casual coffee chat with Anser, NetElixir’s senior search marketer, who commented that Product Listing Ads are performing much better than text ads. Deb, another senior colleague, endorsed Anser by quoting that Google PLAs have outperformed text ads in terms of click-through-rate. He was referring to one of the Adobe blogs he follows – Adobe Blog on Digital Marketing. This discussion by Anser and Deb intrigued me and Keith, who were sipping the hot coffee with them. And this triggered a nice brief productive coffee chat on a few Product Listing Ads optimization tactics.
Multiple product targets for super-granularity
The first technique which Anser and Deb explained was to use multiple product targets. They explained the concept of product targets and how enhanced targeting can be beneficial for PLAs. They said that product listing ads directly expose the products of the retailer unlike other ads which depict the websites. PLAs obtain product information like title, image and price from the product feed saved in the Google Merchant Center account and show them candidly to ensure high visibility. Under the ad groups in PLA campaigns, product targets are set which helps to group the products based on feed attributes like id, brand, product_type, condition, adwords_grouping and adwords_labels. This grouping can be as granular as possible to uniquely identify a product or a set of products. This granularity gives control to bid differently on different sets of products. Anser highlighted the importance of using multiple product targets by quoting how they helped him achieve a 15% increase in R/C for one of his clients. Meanwhile, Deb clarified that in PLA campaigns, product-targets play the role of keywords. The search queries match the product-targets and mapped products are advertised upon.
Negative keywords for PLAs
According to Anser, the multiple product-targets tactic has to be complemented by techniques like negative keywords in ad groups. For example, let us assume that the product targets used are “brand=XYZ;product_type=shirt” and if the feed doesn’t carry shirts for kids then we should add the negative keyword: “-kid” to our ad-group and our PLAs will not show for any search queries that contain the word “kid”. Anser observed that this feature has helped him to specifically target the most interested customers with reduced costs and better ROI. For one of his clients in the beauty-care industry, he could reduce the costs by 18% in a period of 2 months using negative keywords.
Using Product Filters
Having heard these interesting tactics made us more fascinated by PLAs. Meanwhile, Keith asked a valid and interesting question – “What if I never want to advertise certain types of products listed in the feed? Let us assume my campaign is exclusively for books while the Google Merchant Center feed features shoes and apparel in addition to books. How can I exclude them and focus only on books?” Deb was quick to answer Keith as he introduced the concept of product filters. Product filters define which products in the feed should appear for your product listing ads. For the problem which Keith described, the right solution would be to use product filters at the campaign level which sets the product_type as books. Once this filter is set, only books are eligible to show on your product listing ads.
Maintaining Updated Feed
Deb didn’t stop at product filters. He brought in a new angle to PLAs by raising a very practical issue -“What if we do everything right on PLAs and drive the user to buy a product which is actually out of stock?” As we became absorbed by the intricacy of the question, Deb vehemently stressed the importance of maintaining a feed that includes the latest product information such as price and availability status. An updated version of the feed would make sure that it matched the store inventory exactly. Deb jokingly said that an out-of-stock product should never be in-stock in the feed and assured that updated feeds would avoid the situation which he described earlier.
The discussion ended with that last point of maintaining an updated feed. By then we had finished our coffees and wanted to rush back to our desks to finish our weekly client reports. However, the discussion left me wondering how a simple coffee chat helped us understand some interesting search marketing tactics. I wished these types of discussions happened more frequently. As we were wrapping up our chat, I, till then silently listening, made a closing remark – “It’s true that a lot can happen over coffee”.
By Subramanian S.