Dynamic Creative Optimization: Changing the focus of the advertising business from creative master minds to data algorithms | Articles

The nature of the advertising creative process we know from Don Draper and his team on Madison Avenue has shifted. Leveraging real-time data signals to tell one-to-one brand stories is changing the advertising business.


There is much ado about winning over the intent-rich moments when potential customers are shaping brand preferences. Programmatic media buying has unlocked tremendous potential in how advertisers can contextualize their brands in these moments. Micro-moments are the new battleground and marketers need their advertising to cut through the noise. Brands are challenged more than ever before to tell their stories with resonant creatives that encompass both data and emotion.

The definition of creativity is changing

Where the traditional approach was to show one generic creative to all, dynamic advertising scales the creation of personalized ads. Yet a prime feature most dynamic advertising campaigns lack is the ability to automate the optimization. Marketers are optimizing their campaigns statistically based on performance data. As such many factors that may play a significant role in the performance of the creative are being ignored.

Dynamic Creative Optimization (DCO), is a display ad technology that enables data and creativity to come together. This approach covers both ad production and creative optimization using real-time technology. DCO uses feeds of data and sets of rules to fuel smarter creatives. The technology can accept a wide array of data signals: from audience data, website analytics data to contextual data such as time of day, device, ad placement, location, weather, etc. The algorithm interprets a set of variables and uses a rules-driven approach to serve tailored designs to each target audience segment.

The solution will automatically swap out different dynamic elements in the ads such as ad copy, offers landing pages, button colors, images and more. Its ability of moving parts against variables allow real-time multi-variate A/B testing. Subsequently the algorithm will interpret the test results and optimize the creative development on its own using engagement KPI’s like a user action or post engagement. This form of programmatic advertising can bring marketers closer to showing their target audiences the most relevant ad at the most appropriate and influential time.

A real case: Bringing to life the signal-driven creative

The following real case illustrates the use of this technology. An advertiser active in the travel industry wanted to boost its digital business by aligning creative messaging with all available audience segments in their DMP. To dictate the most logical audience-creative pairings, the following creative decision tree was deployed. For each impression, the user’s real time information is passed on from the DSP to the ad server. There each user is assigned to a node of the creative decision tree.

The outcome? The number of display ads created by DCO was mind numbing: 7.000 banner variations. As a result click-through rate increased by 29%, conversion rate by 195% and average order value by 5%.

The Downside of the new creative paradigm

David Moore, chairman of WPP, says in a interview with Beet.TV: “Frankly, I don’t know why everybody isn’t doing it”. Probably because it is not justifiable for everyone. Unfortunately, setting up DCO is a complex and technical process that requires a reasonable buffer time and technical skills. First, marketers need to plot decision waterfalls that will drive the campaign. To configure the automated decision, data needs to be incorporated into the ads by linking the data feeds. Finally, DMP audiences need to be synced with DSP line items, multiple ad tags need to be created and trafficked to each line item, making the related ad serving costs 3 times higher, than a programmatic creative.

The amount of knowhow and set-up required makes DCO a better fit for advertisers with large campaigns where even small gains in performance loom large on the bottom line. The solution might interest marketers managing direct response campaigns with millions of impressions budgeted that allow multi-variate testing and machine-learning.

So, who is in the driver’s seat Madison Avenue or Silicon Valley?

In the modern digital advertising ecosystem data has taken a bigger part of the data-emotion equation. Advertisers are challenged to recruit a sophisticated team of specialists to make it work. So far, a study done by PaperG and AppNexus indicated that 97% of the campaigns do not serve unique creatives for each placement. DCO might have shaken up the ad creative world, but advertisers are pulling in their horns and are becoming very risk averse.

Author: Dhan Claes



Get in touch

Semetis | 44 Rue des Palais, 1030 Brussels - Belgium

+32 (0)2 211 34 50


Connect with us

Cookie Policy

This website uses cookies that are necessary to its functioning and required to achieve the purposes illustrated in the privacy policy. By accepting this OR scrolling this page OR continuing to browse, you agree to our privacy policy.