April 16, 2013
You’ve asked for it, and here it is: The Total Guide To Product Listing Ads. We have a feeling this product will continue to expand and get cool, new features since it’s doing so well for our eCommerce accounts. But, there is so little information out there about how to set these up and make them work for you right now. We took it upon ourselves to put everything we know about PLAs into one, six part guide. We’ll be posting twice a week until all six parts are posted, so keep your eyes peeled!
Part 1: Setting Up Your Data Feed
Part 2: How to Set Up Your PLA Campaign
Part 3: How To Optimize Your PLA Campaign In Your Data Feed
Part 4: How to optimize your PLA campaign in AdWords
Part 5: Targeting Strategies
Part 6: Using Google Analytics To Measure Performance
Part 5: Targeting Strategies
The most important thing, which I will reiterate when discussing targeting, is to target by the difference that makes performance change. For instance, if you find performance is about the same for each brand of dog food you sell, but each brand name is different from all of your other brand names, then you should target by this attribute.
I’ll go through some of the more successful and most commonly useful targeting strategies to begin with, and hopefully you’ll find one that matches your needs or could be altered slightly to match them.
If you sell items that have different brands, and you see different performance by brand, this is a great option for you. You will want to set up an ad group for each brand, and within that ad group, set up a target for each brand. Start with a safe minimum bid. I normally judge this based on what I’m comfortable paying in my Search campaigns. If my average cost-per-click is $.75 in my Search campaigns, I will start with that as my Max CPC in new programs like Product Listing Ads. During the Analysis section, I’ll focus on how to tell you need to change your bids in these campaigns.
- Set up targets by product type.
If you sell many different product types, and you see different performance based on these product types, this is an excellent choice for you. The set up is the same as the targeting by brand: set up ad groups based on product types. Within those ad groups, you should set up targets based on product types. When I say product types, I am referring to the Google product categories.
Example Google product categories:
Animals & Pet Supplies > Pet Supplies > Bird Supplies > Bird Food
Animals & Pet Supplies > Pet Supplies > Bird Supplies > Bird Treats
Animals & Pet Supplies > Pet Supplies > Cat Supplies > Cat Apparel
This attribute is not required in your data feed when setting up your initial data feed, but it is useful for optimizing your campaigns. I would strongly suggest taking the time to go through your product to find the most specific product category for each product.
- Set up targets by custom defined attributes.
AdWords labels targeting and AdWords grouping targeting allows you to target by whatever grouping you find relevant. This would allow you to group things by season, for instance. You could have labels for “fall line” and “summer line”. You could even include a designer’s name: “Gucci Fall Line”. The possibilities are endless here. The key to utilizing this targeting method most effectively is to find the biggest cause of performance change in your account. You need to analyze your data in AdWords and any Analytics packages you use to find what attribute you can separate your products by that cause performance changes, but everything within each attribute has similar performance. So, while these targeting choices are the most appealing because they allow the most customization, they also require the most work and intelligent analysis to utilize properly.
There are a few differences to these targeting methods you should consider before picking on. You must use the cost-per-click pricing model with the AdWords labels target, but you can target up to 10 labels at a time. Whereas with the AdWords grouping target, you can only target one grouping at a time, but you can use it for whatever bidding method you prefer.
- Set up targets by SKU or product id.
This is a targeting method that seems favored by many advertisers, as I’ve learned from discussing Product Listing Ads on blogs and social networking sites. This allows you to know exactly what product you are targeting. Instead of grouping several or many products into one target, you can be sure the performance of each target is directly tied to an individual product.
This strategy could be advantageous to any website with a small inventory, with a handful of top performing products, or with an account manager that has a great memory! For example, if you sell a small line of supplements, it may benefit you to manage each product separately in your Product Listing Ad campaign.
As to the organization of this targeting method, you could either have an ad group for each target, which would be beneficial if you’re doing cost-per-acquisition bidding at the ad group level. You could also think of some other way to organize your products, perhaps an ad group for each product type, and then have the individual product targets housed within each ad group. This would be best for cost-per-click bidding.
Analysis of targeting strategies
As I covered earlier in this article, the key factor to a good strategy for this product is to determine the key factor that changes performance, yet has similar performance within the segmentation. Hopefully you were able to perform an in-depth and intelligent analysis to find what this is in your account. However, you’ll want to make sure that you’ve found the key factors to segment your targeting by through analysis of performance.
Download a full report from AdWords that includes all target performance for as long of a date range as is available and makes sense. If you’ve had the same strategy in place for 30 days, and you’ve seen hundreds of clicks for each target, then I would pull data for 30 days. Just make sure whatever date range you select has had the same strategy for the entirety and that you have enough data gathered to feel confident in the statistical validity.
Once you have this report downloaded, compare performance between targets. Hopefully you see enough differences to justify continuing the segmentations you’ve selected as your targeting strategy. If you see some similar segments have very similar performance, you may want to consider grouping these into one target for easier management.
Comparing keyword-triggering data is also vital in analyzing your strategy. Performance comparison at the targeting level could be misleading if you haven’t set up good negative keyword lists to limit cross-contamination between targets. Make sure keywords are being sent to their most relevant target. You may need to perform analysis here and continue adding negative keywords to ad groups to prevent cross traffic several times before you can analyze your target strategy properly. Make sure you allow enough data to accrue after the last time you add negatives so that you have clean, statistically relevant data to analyze.