The previous post discussed best practices and strategy for developing awesome audience lists via Adwords Remarketing and Google Analytics. In this post we’ll discuss bidding strategy and how lists can be combined with other Google Display Network targeting options for hyper-relevant ad serving.
As with any instance in which AdWords is given the opportunity, the system will prioritize the content with the highest bid that’s eligible to serve. This means that if you have a user who fits into multiple audience lists, and those lists are all given their own ad group, the ad group with the highest bid will get the impression. For example, let’s say you have just two lists — homepage viewers and product page viewers. Presumably, many users qualify for both lists because many of the product page viewers entered the site on the homepage and navigated to the product pages from there. If the “homepage viewer” ad group has a higher bid, those are the ads both buckets will see, because they both qualify for the homepage list.
It makes little sense that you would bid higher on your homepage viewers list than on the more engaged list, but there are many instances in which it’s not so clear cut, or when bids start being based on statistically insignificant data and not assumptions about which list is more likely to convert.
For this reason, be sure you have a bidding strategy that makes sense, and create lists with sufficient exclusions to keep your buckets isolated. (See my last post for tips on how to do this.)
If you’re targeting the same bucket of users, but segmented by recency, you should start with a tier-based bidding system. This is the most effective way to ensure maximum bid flexibility. For example, let’s say you’re targeting cart abandoners in three sub-segments:
Cart Abandoners: 1-3 days
Cart Abandoners: 4-15 days
Cart Abandoners: 16-30 days
Presumably, there’s a correlation between a user’s likeliness to convert and the recency with which they added their product to the cart. This should be reflected in your bidding:
Cart Abandoners: 1-3 days = $3 CPC (highest bid)
Cart Abandoners: 4-15 days = $2 CPC (highest bid)
Cart Abandoners: 16-30 days = $1 CPC (highest bid)
Don’t forget to create negative lists for these sub-segments as well!
Combining Targeting Options
Remarketing used to be more isolated from the rest of the GDN targeting options. Nowadays, audience lists can be added as easily as keywords or managed placements to Display campaigns. This flexibility is super useful because it allows you to create extremely tight targeting.
How you might use this:
- Bolster more general audience lists (homepage viewers only) by adding custom placements so your ads only show on sites relevant to yours. This helps ensure that these users are in a relevant frame of mind when they reach your site.
- Add Topics and Interest targeting to ensure serving to your past users who’ve also demonstrated in interest in relevant topics only while they’re viewing relevant placements.
- Add contextual keyword targeting to ensure relevance of placements without having to hand pick the sites on which you want to serve.
Set List Durations
You can also set the length of time before AdWords drops a user from a Remarketing list. For example, a list duration of thirty days would only include users who visited your site within the past thirty days. This can make a big difference in the performance of your campaigns because there is often a correlation between the recency of a user’s visit to your site and the likelihood they’ll convert.
Setting Frequency Capping
This is one of the most important settings you’ll make, and how you use it will determine whether your ads are useful to users or brand killing nuisances.
Frequency Capping allows you to set the number of times per day, week, or month your Remarketing ads are eligible to appear for each unique user. Keep in mind that this applies to each unique user within each individual campaign so if you have more than one campaign targeting the same user, you’ll want to spread your target number of impressions over multiple campaigns. For example, if you have three campaigns eligible to serve to the same user and you want that user to see your remarketing as a maximum of nine times per week, you’ll want to set each campaign’s Frequency Cap to three.
Once you have some data, you can view your reach and frequency stats in the Dimensions Report to determine the ideal impression frequency for your ads.
Top Five Optimizations
- Display Network Lost Impression Share Report – This report will show you how much exposure you’re missing out on due to your bids. If performance justifies it, experiment with raising your bids for increased impression share.
- Landing Page Testing – As with any type of campaign, the destination of your ads has a direct effect on conversion rates. Experiment with different landing pages and always make sure the pages you do choose are directly relevant to the content of your ads.
- Mine Placement Reports – As with any Display Campaign, you can see a list of the sites on which your ads appeared. Pay attention to which sites are performing well and bid down on or exclude those that aren’t.
- Test Different Frequency Caps – As we went over before, the frequency with which users see your ads has a great deal to do with how they will perform. Experiment with different frequency levels to find your sweet spot.
- Testing Different Ad Creative – Try to run multiple ads against each other and be sure to test your text ads against your image ads.
- More list granularity means more bidding flexibility. Segment your lists to the greatest degree that’s practical and have that segmentation reflected in your campaign structure.
- Use audience lists in combination with other GDM targeting options
- Use tier-based bidding when segmenting audience lists based on recency
- Frequency caps to control how often users are eligible to see your ads
- Set membership list durations to control recency