The Oracle Data Cloud blog highlights the latest data-driven insights and trends in digital marketing and ad tech.

Managing advertising frequency: 3 Tips for holiday success

This week’s guest blog post is contributed by Jim Greco, Senior Analytics Director, Oracle Data Cloud.

For those of us in the digital marketing industry, it’s a well-known fact the last quarter of the year is the most active and competitive for available advertising inventory. 

According to eMarketer, fourth-quarter spending in 2016 was more than $650 billion and ad spending grew +64 percent versus the same period in 2016. This is a noticeable difference from the 40-50 percent growth rates in other quarters. 

Simultaneously, prices charged for Q4 advertising also are increasing. eMarketer reports a CPM growth of roughly +20 percent versus Q4 2015—and figures from Oracle Data Cloud suggest it could be even higher.

For many advertisers, Q4 and holiday business performance can spell the difference between a successful or unsuccessful year. 

At Oracle Data Cloud, we believe one of the most practical steps an advertiser can take to make their holiday ad-spending productive is to avoid overserving impressions to the same households in their campaigns. 

Despite the promise of complex optimization algorithms, programmatic campaigns frequently exhibit skewed impression distributions and wasteful spending. 

We explored 34 recent programmatic campaigns for CPG advertisers and found widespread evidence of cookie overserving.   

Here are some stats that may surprise you:

400K – The most impressions served to a single cookie within the 34 CPG Open Web campaigns included in the analysis.

36% – The average amount of campaign budget spent on top five percent of exposed cookies.

$198K – The average media spend on the top five percent of the exposed audience across all 34 studies included in the analysis.

$5.2MM – The amount of money saved (out of the total of ~$14MM spent) if the frequency of those top five percent of cookies were better controlled for.

If you’re disturbed by these stats, then consider taking a few practical steps to make your Thanksgiving and Holiday campaigns more cheerful.

1) Talk to your agency and publisher partners about the availability of mechanisms to “frequency cap.”

Apply a maximum limit to the number of impressions a cookie or user can see. It’s OK to choose a number that assumes multiple exposures during the campaign duration, since the goal is to avoid the previously mentioned five percent phenomenon.

2) Avoid running short-duration campaigns.

By concentrating spend into a shorter campaign window, you’ll be prone to inflating frequency because people simply cannot visit the same number and diversity of sites as they do in a longer campaign.

3) Avoid overly narrow targeting

This is a tricky one. Targeting certain consumers can be very effective at improving your ad spend. But if that target is too narrow, you’ll end up serving algorithms deciding they can only find and serve the same consumers repeatedly because you’ve limited their choices.

By taking these steps, we believe the likelihood is good you’ll enjoy a prosperous and successful holiday advertising season!

We’re also here to provide one-on-one help, so contact The Data Hotline today with your questions. (What's The Data Hotline?)

About Jim Greco

Jim is Senior Analytics Director at Oracle Data Cloud, dedicating his career to translating the sophistication and power of data science into real-world applications that improve the advertising and marketing performance of consumer brands.

His current work with Oracle Data Cloud focuses on helping consumer goods advertisers execute successful digital marketing campaigns on the world’s largest media platforms such as Facebook, YouTube, Twitter, Pinterest, Amazon and Snapchat.

Stay up to date with all the latest in data-driven news by following @OracleDataCloud on Twitter and Facebook!

Image: Shutterstock  

Be the first to comment

Comments ( 0 )
Please enter your name.Please provide a valid email address.Please enter a comment.CAPTCHA challenge response provided was incorrect. Please try again.