I came across some interesting research from DoubleClick that compares the performance of online video advertisements (in-page video ads as opposed to in-stream video ads) with that of traditional, or image banner ads (GIFs & JPEGs).  More and more, I am seeing videos run by themselves in banner positions (mostly 350*250 medium rectangle) on websites - hence….  "video banner ads”.

Which is More Effective?

The question seems to be an easy one to answer; video banner advertisements far outperform that of traditional, or image banner ads. According to the study, videos run as advertisements received CTR (click through ratios) between 4 and 7 times that of image banner ads.  Actual click-thru rates (CTR) for video ads were measured anywhere from .4% to .7% whereas GIF and JPEG ads only averaged about .1% CTR.

What about Interaction?

But CTR, would only be one way to look at this.  When comparing click interactions, it is important to realize that with video banner ads, one can actually measure the length of interaction as well, or the duration of video watched.  And, just as we have seen with in-stream video advertisements, users tend to watch more than 2/3rd of the video on average.   So, not only are users interacting more often with video advertisements, but they are captivated by the advertising message for a measurable duration.

Although the research does not measure additional interactions within the video advertisement, one additional advantage to video ads is that you can potentially track conversions.  With a video (depending on the player used), one could overlay links to the advertiser's website, email contact forms, and other methods for conversion.  Imagine the power of telling an advertiser that not only did their ad receive a .7% CTR, but that X percentage of viewers converted, and X leads were generated.

Conclusions and Lessons Learned - New Research Needed

Knowing that consumers watch 66% of video ads on average, advertisers should do their best to create a strong message that occurs early within the video.  This study ran only video advertisements that were 15 and 30 sec. in duration.  It would be interesting to measure the effect of running longer video ads within banner position although currently IAB guidelines say no more than 15sec - We'll talk about that later.

It is important to note that it is important that when running video banner ads, your player should have a prominent "PLAY" button. Each of the video banner ads analyzed in the DoubleClick study had play, pause and stop buttons.  In addition, DoubleClick measured which of the 3 buttons were most often used and found that the video control consumers clicked on most was the "play" button.  In essence, the play button functions as the video banner ad's subliminal "call to action.”

It is also important to note that the majority of the videos in the research study were auto-play and audio was user-initiated.  I wonder if the same results would be realized if the video were also user-initiated. This would lend even more weight to the idea that the "play" button has some special effect on users.

Finally, it would be interesting to see this research updated with a comparison of video banner ads versus flash based animated banner advertisements.  If videos were to outperform flash ads as well, this could also add to the idea that the increase in performance is heavily related to the instance of a "play" control.

If anyone is aware of any additional research with regard to this, please let us know in the comments section below.