What happens if you let loose a few brilliant Masters students to find empirical answers to some of the ad industry's most burning questions? 
Maybe you get nothing -but maybe some of them come back to you with answers that whilst may not be the holy grail in advertising but nagging enough to question some of the ways the industry is currently thinking. This is just the experiment that I'm currently setting up between DDB UK and my psychology department at Goldsmiths, University of London. 
The idea is that a few of our current Masters students work on projects that are of high interest to the advertising industry but apply all the academic rigour and standards of empirical testing to ensure they also get their MSc degree in the end. 
DDB came over to Goldsmiths this week to present themselves and the questions for which they need answers to students and the distinguished academic staff. Students were absolutely excited and even the long-term academic staff was intrigued by the glimpse into the funny world of advertising. I expected many more questions of the type 'aren't you all evil manipulators?' from this clash of thinking cultures. But either their good education or the genuine surprise that advertisers can be brilliant practioners in perception and cognition but without having any clue about academic psychology made for an open discussion but with a very positive vibe. 
Ok, now what are the burning questions that DDB thinks are the secret to advertising success? 
Well one of the hottest topics in these days of tight budgets is the virality of ads. If you can get ordinary people to distribute your ad for you, then your client could save a lot on media budgets (see why I thought the evil manipulators stereotype would come up in the discussion?). We know that people love to share in general, be it music playlists, youtube videos, cartoons, spoof powerpoint presentations (although I haven't seen one of those for a while) etc. But why do they share one video or a particular TV ad and not the next one? Of course advertisers and media professionals have asked that question before and of course people have come up with long lists of post-hoc answers (explaining why one particular ad has been shared a lot is easy after the fact). Now, the Masters project will look at different aspects that are relevant here, psychological properties of the individual sharer, the content of the ad, and the context in which the ad is shared. The challenge for the student and his/her supervisors from Goldsmiths and DDB is to come up with a predictive model that allows to forecast the rate of sharing before the ad is actually released: the buzz index (it's worth nothing without a fancy name - after all we are in advertising). 
Clients and advertisers join in for regular pre-testing rituals of ads using tests that haven't changed in a long while. Project number two asks whether the usual techniques of pre-testing in advertising are missing out on something. Psychologists would classify the mainstream methods of ad pre-testing as explicit behavioural tests that rely on the participants' ability to give an overt and bias-free account of the perceptual and emotional processes that an ad stimulates in a typical pre-testing session (I really do love academic language). In psychological research a revolution discovering the importance of implicit processing and memory started more than 35 years ago but none of the many and very powerful results from this research have made their way into the mainstream of ad pre-testing. Potential consequences from this implicit neglect are obvious: Mainstream pre-testing that ignores implicit cognition can favour ads that aren't effective in the real world ('false positives', as we call them in signal detection theory) and can also reject ads that would work brilliantly but never make it beyond pre-testing ('misses'). Now, can we identify certain types of ads that prove their effectiveness only in the domain of implicit cognition and where the current standards of advertising are just not appropriate? Our masters student will tell us in a few months time. 
When it comes to implicit and unpredictable effects in advertising music (in TV ads) is a very strong candidate. The discrepancy between the huge effects that the right music can have on an ad and how little we know about why music works is striking. So many effective ads get their punch from the music that goes with them (possibly far more so than from any "message") and creatives know very well how to set off the right emotions in us with just the right song. But still, music is vastly undervalued in advertising, probably because no one really knows how it works (and we tend to value things much more that we understand). This is a call to a Scientist in Residence with a music background! Project number three is designed to quantify in empirical terms how much more effectiveness an ad can gain with the right music but also how much you can lose when the music is not right. And music is just the right vehicle to investigate the power of implicit processing: Most people can't name the chords of that sweet sequence of harmonies they are hearing and they can't tell that it is the rising major sixth just in the right place that is responsible for the big lump in their throat. Music works, even (or maybe particularly) for those people that have never seen a music school from the inside.