By this point in the evolution of digital marketing, we’ve all gotten a glimpse of retargeting programs “in the wild.” My first experience was when I had browsed Zappos looking for a gym bag and I stumbled upon a bright pink Nike duffle. Perhaps I spent too much time looking at the bag. Maybe I visited that page more than once, or maybe I just accidentally clicked into the product to bounce right back to the results page. Whatever the case, that pink Nike duffle bag refused to quit, following me around the web for a number of months, and unfortunately for Zappos, my interest in heading to the gym doesn’t seem to last more than a couple of days.
Although multiple years have passed since that pink bag stalking, I still see companies struggle when implementing retargeting campaigns. Frequently, I’m asked about what can be done to improve the end user’s experience or how investment can be optimized to improve return. Here are some of the common problems reported and recommended solutions:
“I see my ad everywhere, on practically every page load, and sometimes multiple times! It seems like we are overdoing it.”
This is one of the most common concerns about retargeting I’ve heard voiced by marketers and is amplified by the number of auctions that occur on a single page. Unlike paid search, where a single auction occurs, in the world of display, each page load can have several independent auctions happening simultaneously. To minimize or eliminate this problem, several steps are recommended.
1. Avoid running retargeting through multiple platforms at the same time.
Sure, it seems like a good idea to set up a bake-off between retargeting platforms, but this fundamental configuration sets you up for reduced control (and less-than-optimal efficiency), which causes the above problem.
When two separate retargeting programs are run concurrently, the frequency of how often a user is exposed to your ad is less controllable. One platform may have a frequency cap of three impressions per day and the other may have a cap of three per day, too. Since they aren’t communicating, that user can be exposed to six impressions per day. By running your remarketing through a single platform, you will take one step towards reducing undesirably high frequencies.
2. Use a system that allows for multiple levels of frequency caps.
As retargeting technology and audience strategy have evolved, campaign structure has become more complex. Oftentimes, this complexity is realized through multiple platform line items set to target sub-segments of your retargeting pool. For example, you may be targeting cart abandoners in one line and then retargeting FAQ page visitors under a separate line. If your system supports line item frequency caps, you again could have a frequency cap of three impressions per user on each line, resulting in six impressions per user -– an undesirable result.
However, if you implement a platform that supports frequency caps across the structure hierarchy, you can introduce line item, campaign-level and, in some systems, universal frequency caps, reducing or eliminating this problem completely.
3. Determine an audience priority logic and implement audience suppression.
As in the above example, a user might be a member of several different retargeting audiences. They could be in the homepage visitor, FAQ page visitor and cart abandoner audiences all at the same time. They could also be part of audiences defined by how recently the action occurred, for example, audiences based on activity in the last one, seven or 30 days.
If you’ve gone to the trouble of segmenting your audiences in this way, the likelihood that you have a campaign structure to treat these audiences differently is probably high. Minimize the chance of over-serving an individual by developing a targeting priority or waterfall, suppressing audiences accordingly.
4. Finally, use a system that supports frequency caps over the smallest period of time possible.
To minimize seeing multiple impressions on a single page load, make sure you’ve followed the above three tips. Beyond that, run your remarketing through a system that provides frequency cap functionality at the smallest time increments possible. While the ideal would truly be less-than-millisecond frequency caps to account for the load speed and multiple auctions taking place on a single page, look for minutes or seconds, if available.
“I already purchased that product; I don’t need that product anymore. Why is it still following me?”
This is another classic challenge that advertisers face. At several industry conferences, I’ve heard speakers or panels publicly call out marketing campaigns where this sub-optimal user experience runs rampant. Here are a few ways to prevent and minimize this problem:
1. Activate your purchase data and suppress.
This is certainly the most obvious action to take. If a user has made a purchase on your site or in your store, develop custom audiences from web activity and onboard offline data sets to suppress buyers from your program. Dependent on your business model, however, you may consider altering this straightforward approach.
In the case of retail and e-commerce advertisers with multiple SKUs, you may want to collect SKU-level data of buyers. This would enable you to stop showing the previously-purchased product, but continue to target users with complementary or recommended products.
In the case of lead generation programs, like in enterprise B2B sales, you may want to suppress lead form completions from lead-focused retargeting programs while intentionally setting up separate campaigns to nurture with alternative proof points, case studies and calls-to-action.
2. Implement more stringent retargeting durations.
If a user viewed a product 30 days ago, think about whether it is appropriate to serve them a message with that specific SKU. If I looked at a gym bag 30 days ago and you’ve already served me 30+ impressions showing that product, isn’t there a point where you might assume I’m no longer interested in buying? Alternatively, I may have purchased elsewhere and no longer have a need for your product or service.
Step back and think about this from the consumer perspective to qualitatively evaluate. Leverage your frequency, time lag and performance data to quantitatively evaluate. Once you’ve assessed both sides, put a plan in place to alter your targeting duration and messaging.
“I’ve seen multiple ads with two different offers. It’s confusing and it’s not what I intended.”
Regardless of whether you’ve taken the steps above, if you aren’t thinking about your program holistically, you may be leaving yourself vulnerable to this sub-optimal user experience and reduced program efficiency. How does retargeting fit into your larger digital marketing program?
1. Consolidate and suppress.
Similar to the retargeting platform consolidation discussed above, programs that are managed in isolation of each other have no way of communicating to reduce or eliminate the overlap. A user can fluidly move from being eligible for a prospecting auction to a retargeting auction, but their experience with your brand is not divided by rows on a budget spreadsheet.
By consolidating audience strategy, either by working with a sophisticated audience-focused partner and/or by investing in a data management, campaigns can be configured more effectively to reduce overlap for a single user. Suppressing retargeted audiences from prospecting campaigns will allow for greater control over the messages served to your audience and improve how funds are allocated to maximize program return.
Overall, the introduction and evolution of retargeting has equipped marketers with several dynamic and flexible options to reach previously engaged users. It is up to us as marketers to leverage the technology more effectively to maximize program return, and more importantly, eliminate historically unfavorable and intrusive practices to provide prospects with more relevant and meaningful brand experiences.