An active comment thread has sprung up on Young's recent post about branding metrics and the web. One of the topics (Pepsi versus Coke) has recently had some ground-breaking research come out regarding how branding actually works in the brain.

This month some of the analysts in Microsoft Advertising are reading Predictably Irrational by Dan Ariely (quick review: if you liked Freakonomics and you read all of Gladwell's stuff, it's probably worth your time). The book highlights the well-known "soda wars" and the taste test results between Coke and Pepsi. (Full disclosure: I grew up in Atlanta where Coke is the second-coming and Pepsi is the fruit of the devil.) Both drink manufacturers have done taste tests that "show" their product is superior. How is that possible? Pepsi does their taste tests blind, Coke does them with brands revealed. As Ariely asks on page 166 of Predictably Irrational, "Could it be that Pepsi tasted better in a blind taste test but that Coke tasted better in a non-blind test?"

Ariely details the research of a group of neuroscientists who hooked taste testers up to an fMRI to see what was going on in the brain. (Their article can be found here.) In Ariely's word, "as the participants received a drink, they were also presented with visual information indicating either that Coke was coming, that Pepsi was coming, or that an unknown drink was coming." And the researchers replicated the cola war results-what happened in people's brains depended on whether or not the brand was shown.

Tasting the drinks triggered one part of the brain (yeah, I know you're curious, it was the ventromedial prefrontal cortex - VMPFC - of course). But if people saw the brand before the drink there was another part of the brain stimulated (naturally the dorsolateral aspect of the prefrontal cortex or DLPFC). So tasting some sugar water fires up the VMPFC, seeing a brand engages the DLPFC. Are you with me?

What's interesting is the sugar-water response is roughly similar, but the brand response is much greater for Coke. As Ariely says, "this should be good news to any ad agency, of course, because it means that the bright red can, swirling script, and the myriad messages that have come down to consumers over the years (such as "Things go better with.") are as much responsible for our love of Coke as the brown bubbly stuff itself."

Well, first of all, I'd say that yes, Dan, those of us in advertising are excited to see the results. If you get a nice shot of dopamine from branding and that adds to your enjoyment, then all of the sudden we don't have this hierarchy of "value" with intrinsic quality above perceived quality. I've always naively assumed that it's more important for a product to be of high quality than for it to have a good brand. (Yeah, yeah, I know that the last 100 years of retail from Hermes right through Old Navy contradicts that.) I still think their result is interesting and counter-intuitive. Could a product test this and look at the ratio (maybe normalized) of activation between VMPFC and DLPFC and figure out if they need to put more money into product R&D or marketing? Put another way, if your product is firing lots of VMPFC (sugar-water) but little DLPFC (brand), you need to put some money into marketing.

Second, how does this branding effect happen? If you grow up in Atlanta and every public space is branded with the Coke logo then every neuron of your reptile brain is probably going to fire every time you see the Coke logo. But what if you're Expedia? How do you create this brand experience so that in the cluttered space of travel reservation your brand elicits a greater response in our brains so that someone books through you instead of through Orbitz? Or is the concept of brand for intangible products different?

Finally, what does this mean in the short term for marketers? Obviously we don't know the answer-I certainly don't. But I'd be willing to bet that the ultimate equation is going to look something like this:

Brand Impact = Audience + Context + Duration + Media + Creative

· Audience: You're not selling teenagers on life insurance, you're not selling someone who's never left Minnesota a cruise. You're just not. Of course, this isn't to say you can brand someone before they're ready for the product. Let the Lexus-sponsored cartoons roll..

· Context: When you are talking to your audience matters. When I'm on a business trip in an unfamiliar city, I'd pay to see one minute VOD ads on local restaurants. When I'm trying to find a link to the right article in the journal Neuron, it's pretty frickin' annoying. Context clearly matters.

· Duration: It's obvious that more duration is better, as long as the audience is controlling the interaction. In the final accounting of brand impact the role of duration will be a critical underpinning. (Incidentally, there'll be a future blog post tying together the ideas of creative, context, and duration.)

· Media: The gold standard is probably an HD long-form VOD commercial on a big screen TV, though there are arguments. What about when Target takes over The New Yorker? Pretty impressive use of the medium, but we don't know how to order it yet.

· Creative: I'm not a creative guy, but we certainly see big spreads in response online across creative when we control for the above factors. Creative can make or break a campaign.

What am I missing from this list? To me the interesting open questions right now are figuring out how to paste these together. If you're selling to males 25-45 would you rather have a prime time broadcast TV 30-second spot going out to 15M households or an online video buy targeted to reach 10M of your audience? Of course you can't answer that without knowing the cost, but we don't even have the tools to start evaluating the answer.

Final question: Does anyone have a portable fMRI that I can borrow for a few years?

Thanks

John Chandler-Pepelnjak

More from Microsoft Atlas Institute