Introduction
Recently, a top journalist confided that he primarily thinks of real-time bidding “as a way to do direct-response campaigns in display.” This is an often-heard refrain, despite many recent launches in the video exchange and platform space (i.e. BRX on the supply side). Why? Newness probably plays a role, but a deeper reason might be that much of the real-time bidding space is often measured and judged by direct-response metrics, such as clicks.
While clicks and impressions make sense for direct-response, many brand marketers buying video obviously care more about the medium’s ability to raise awareness and persuade a target audience, goals for which clicks are a flawed proxy at best according to recent research by Nielsen.
The question then arises: can real-time buying in video advertising deliver on brand metrics or is it all just a bunch of direct-response hot air?
Methodology
Leveraging its partnership with leading survey vendors, TubeMogul executed brand surveys on five video advertising campaigns totaling 44.7 million impressions and 4,487 random survey-takers. Four of the five brands are from Fortune 500 companies, and the list reflects the automotive, consumer packaged good (CPG), insurance and restaurant industries. In terms of format, only 0:15 and 0:30-second pre-roll ads bought via real-time bidding on TubeMogul’s platform were included (more on the inventory TubeMogul aggregates here).
No 3rd-party audience segment data was used; campaigns only leveraged contextual targeting.
Once exposed to a pre-roll ad, respondents were re-targeted with in-banner surveys within 30 days and were not given any direct incentives to answer. To calculate brand lift, a control group was asked the same set of questions and the raw difference between the percent of viewers responding positively was averaged across campaigns.
Possible limitations include standard objections to brand surveys, such as the purported selection bias in the type of people who take surveys. This was mitigated by running especially large samples -- and budgets -- to chase people that were less likely to fill out a survey.
Results / Analysis
Overall, all campaigns show positive lift across questions. Inset is the average across all campaigns:
Intuitively, purchase consideration and intent metrics tend to increase by the number of times someone is exposed to an ad, while click-through rates often peak at the first viewing.
Based on these results, TubeMogul is readying a deeper dive into optimal frequency across formats, industry categories and metrics.
The data is robust enough to run statistical correlations to be able to accurately predict average lift a marketer might expect when they run a video campaign. Both correlations are significant to 0.05.
- Message Assocation: Each additional exposure of a video ad is likely to lift message association by an average of 3.46% over control, with diminishing returns kicking in at six.
- Purchase Consideration / Intent: Each additional exposure of a video ad is likely to lift purchase consideration by an average of 1.67% over control, with diminishing returns similarly kicking in at six.
Please contact TubeMogul if interested in more detailed statistical analysis on these results.
Conclusions
The story the data tells is clear: brands are seeing quantifiable lift from video ads bought from leading exchanges and platforms, with the most pronounced results in purchase consideration.
Interestingly, although “audience buying” is widely considered one of the main benefits of using an exchange or real-time bidding platform, brands are seeing lift even without using 3rd-party audience segments -- only contextual targeting and TubeMogul’s optimization technology were used in the sample campaigns. Also, as others point out, optimizing for clicks appears to actually minimize the potential purchase intent of any given viewer, so to judge a video platform based on clicks is misleading at best.
Of course, many factors are obviously at play in any advertising campaign -- especially in video -- and creative can always impact these numbers (more on that in a subsequent report).
Leveraging its partnership with leading survey vendors, TubeMogul executed brand surveys on five video advertising campaigns totaling 44.7 million impressions and 4,487 random survey-takers. Four of the five brands are from Fortune 500 companies, and the list reflects the automotive, consumer packaged good (CPG), insurance and restaurant industries. In terms of format, only 0:15 and 0:30-second pre-roll ads bought via real-time bidding on TubeMogul’s platform were included (more on the inventory TubeMogul aggregates here).
No 3rd-party audience segment data was used; campaigns only leveraged contextual targeting.
Once exposed to a pre-roll ad, respondents were re-targeted with in-banner surveys within 30 days and were not given any direct incentives to answer. To calculate brand lift, a control group was asked the same set of questions and the raw difference between the percent of viewers responding positively was averaged across campaigns.
Possible limitations include standard objections to brand surveys, such as the purported selection bias in the type of people who take surveys. This was mitigated by running especially large samples -- and budgets -- to chase people that were less likely to fill out a survey.
Results / Analysis
Overall, all campaigns show positive lift across questions. Inset is the average across all campaigns:
Intuitively, purchase consideration and intent metrics tend to increase by the number of times someone is exposed to an ad, while click-through rates often peak at the first viewing.
Based on these results, TubeMogul is readying a deeper dive into optimal frequency across formats, industry categories and metrics.
The data is robust enough to run statistical correlations to be able to accurately predict average lift a marketer might expect when they run a video campaign. Both correlations are significant to 0.05.
- Message Assocation: Each additional exposure of a video ad is likely to lift message association by an average of 3.46% over control, with diminishing returns kicking in at six.
- Purchase Consideration / Intent: Each additional exposure of a video ad is likely to lift purchase consideration by an average of 1.67% over control, with diminishing returns similarly kicking in at six.
Please contact TubeMogul if interested in more detailed statistical analysis on these results.
Conclusions
The story the data tells is clear: brands are seeing quantifiable lift from video ads bought from leading exchanges and platforms, with the most pronounced results in purchase consideration.
Interestingly, although “audience buying” is widely considered one of the main benefits of using an exchange or real-time bidding platform, brands are seeing lift even without using 3rd-party audience segments -- only contextual targeting and TubeMogul’s optimization technology were used in the sample campaigns. Also, as others point out, optimizing for clicks appears to actually minimize the potential purchase intent of any given viewer, so to judge a video platform based on clicks is misleading at best.
Of course, many factors are obviously at play in any advertising campaign -- especially in video -- and creative can always impact these numbers (more on that in a subsequent report).
The story the data tells is clear: brands are seeing quantifiable lift from video ads bought from leading exchanges and platforms, with the most pronounced results in purchase consideration.
Interestingly, although “audience buying” is widely considered one of the main benefits of using an exchange or real-time bidding platform, brands are seeing lift even without using 3rd-party audience segments -- only contextual targeting and TubeMogul’s optimization technology were used in the sample campaigns. Also, as others point out, optimizing for clicks appears to actually minimize the potential purchase intent of any given viewer, so to judge a video platform based on clicks is misleading at best.
Of course, many factors are obviously at play in any advertising campaign -- especially in video -- and creative can always impact these numbers (more on that in a subsequent report).
