Demystifying Tastes in the
Digital Age

Taste is a relative phenomenon. More often than not, we like what we like. From the kind of music we have on our playlist, to the books we read or the movies we stream, to the food we root for—we are spoilt for choices. As intriguing as it may be, the digital age has also added to this indulgence, giving consumers a vast inventory to choose from and an unprecedented amount of data for businesses to understand human choice.

Tom Vanderbilt in his latest book, You May Also Like: Taste in an Age of Endless Choice, demystifies the different dimensions of taste and choice in the digital age, especially in times when there is little chance that we will like tomorrow what we liked today, and even lesser chances of recalling what led us to our earlier choices. He believes that underpinning this complex dynamics of taste and preferences are our unconscious prejudices, deeply influenced by social affiliations.

Deepak Sharma of Wipro in conversation with Tom Vanderbilt on how tastes and preferences go beyond evolutionary survival needs, why there is no accounting for taste, if predicting changing tastes will get harder with the digital generational shift and much more...

DSYour book, Traffic, used an everyday subject to explore rich insights into human behavior. Did its success surprise you?

TV:Certainly. Perhaps in retrospect, it shouldn’t have—traffic, after all, is something that just about everyone deals with in some form, even outside of the automobile. When people asked what I was working on, I found that the mere mention of the subject sparked heated conversation. Some books start a conversation, whereas others, like mine, frame, tap into and lend scientific weight and data to a conversation that was already happening.

DSIn ‘You may also like’, you have explored likes and dislikes primarily through sensory dimensions–food, music, art. why did you pick these three? Why not, say, books?

TV:That’s a good question! I do have a few mentions of books, like the study that looked at how books that won prizes found their ratings on actually go down afterwards. But, maybe, as an author, books just hit too close to home for me. Or they just seem like a more private, interior taste that seems more inscrutable. Although, these days, Amazon can tell where Kindle readers quit a book in much the same way Netflix knows when people stopped watching a TV series.

DSHow much do the likely explanations for our tastes and preferences go beyond evolutionary survival needs? And how much is explained by areas like culture, status and the way the brain processes complex stimuli?

TV:I think we’ve been on a long march, from survival to culture. Once you’ve more than one reliable food source, once you’ve had achieved basic sustenance, you could begin to discriminate among foods—preferring some and stating that others were out of bounds. In our contemporary market economy, it is striking to see how the dynamics of taste and connoisseurship have blossomed, in two different, but related, ways. The first is that the once rare, upper class tastes have been democratized; almost anyone can get a decent red wine or a cup of coffee. But, on the other hand, goods that were once considered commonplace, like axes or toothpicks, are now offered in very pricey, boutique versions.

DS‘There is no accounting for taste'. you have used this phrase often in the book to explore the complexity of the problem. If accounting for taste is hard AND prediction is harder, then how can businesses predict trends better?

TV:Predicting taste is hard, in part because we, as individuals, don’t often fully take into account how our present tastes are going to shift in the future, as well as what events and things that come between may help shape those tastes. There are a few strong dynamics to be aware of, I think. One is familiarity versus novelty. We like what we know, but we are also attracted to new things. The trick is finding a sweet spot somewhere in between, which is what the famed industrial designer Raymond Loewy described as his ‘MAYA’ principle, for ‘most advanced yet acceptable’. We like new things as long as they are not too new. A product like Apple’s Newton was rejected, I think, less for its own attributes than for the fact that we, as a society, hadn’t yet come to grips with the idea of personal electronic devices. Another strong dynamic revolves around what’s been called ‘optimal distinction’. We, as humans, like to feel like we belong, but we also like to feel different. Hence, products or trends can have a steep adoption curve, but then can also have an almost equally precipitous decline (something actually observed in the popular music charts). Companies do well to offer a variety of options, some of them seemingly innocuous, on products because these allow people to buy the same thing, yet a slightly different version, to help preserve the fragile balance of belonging and identity.

Perdicting Taste is hard. In part because we, as individuals, Don't often fully take into account how our present tastes are going to shift in the future. as well as what events and things that come between may help shape those tastes.

DSDo you think predicting changing tastes will get harder with the digital generational shift?

TV:I think it will get harder. The rise of social media means that we have, as businesses or consumers, more ways to see what other people are doing virtually, in real time. This allows one to gain more of an immediate sense of where taste is headed. But this also gives more people the chance to quickly imitate others, and this imitation—because it can infringe on the sense of distinctiveness that I mentioned earlier—can lead to faster counter-imitation. Keeping up with changing tastes is also a challenge for traditional manufacturers, like the car industry. In the time it has traditionally taken them to design new technology and introduce it into product models, the entire ground might have shifted. For example, you design an elaborate entertainment console, but it turns out that your iPhone does a better job delivering what you want in terms of entertainment and information.

DSWhat is the difference between a trained algorithm that Pandora uses to curate music and the experts who judge beer? Are the lines blurring?

TV:I think the lines are blurring, because of the limitations inherent in each model. One judge hardly has the time to sample every song or beer or chocolate bar. Even if he or she could, personal biases will come into play, however objective he or she tries to be, and his or her taste might not match our taste. On the other hand, an algorithm that rather clumsily selects songs for you out of a vast universe, using only the obvious or rigid criteria—they have a Beatles station, so let’s play them a solo John Lennon—can quickly begin to seem sterile. A service like Spotify’s ‘Discover Weekly’ has gotten a lot better, for example, because it not only serves up a bit of what you already know (and like), but stretches a little beyond to find those things you might like, if you could only hear them. In some ways, people might almost trust an algorithmic recommendation more because it seems ‘neutral’.

DSWith every platform trying to personalize recommendations based on your own past behavior or that of your network, is the room for new exploration in taste shrinking?

TV:There is a danger that we all retreat into our own taste cul de sacs. Sometimes, we find new things we like because we experience them in a completely new setting or without expectation. Think, for example, of how many once obscure songs became accidental ‘hits’ after they were featured in a film soundtrack. People left the theater asking, “What’s that song?” And, turns out, it came out five years ago, but they had never heard it because it didn’t seem to match their taste then. It’s important to keep this genuine sense of discovery available, online or in real life. Chris Anderson, the former editor of Wired, once described it nicely: He didn’t want a recommendation system that told him what his friends were listening to; he wanted one that told him what his friend’s friends were listening to.

DSDoes this discourage brands from breakthrough innovation because it’s too risky? Or can they use some of the ideas in your book to encourage novelty and experimentation?

TV:Because predicting taste is so hard, I think one potential lesson is not to make such big bets. A few years back, I spent some time with Planet Labs, which makes low-cost micro satellites. Their model was meant to oppose the NASA model of spending a lot of money and taking a lot of time to launch big satellites. Instead, they just threw up a bunch of them-so what if some fail, they’ve got more! Clearly, sometimes, companies need NASA-style launches and big products, but as it can be so hard to predict success, it might be better to follow a more agile approach. I was also intrigued by the approach of First Build, a sort of in-house spinoff started by GE in Louisville, Kentucky. Rather than trying to invent things in-house, they solicit a community of people for what products they would like to see. Then they build prototypes in a small-scale ‘maker’ style lab, and fund the product on Kickstarter. By getting out of its normal institutional mode, GE is able to deliver new kinds of innovation with less risk.

DSWith more data everywhere—social media, sensors, mobiles, wearables and Internet of Things — how can we get smarter at understanding tastes and preferences?

TV:One of the key things about so many of these technologies is that they measure behavior as it’s happening. This turns out to give a much more robust picture of people’s actual behavior and preferences than by listening to what they say they like. Netflix has been able to leverage, quite successfully, its own internal data about people’s preferences—based on their viewing behavior—to create new content. Of course, there can even be a risk in that—if you create new things based on what people have liked in the past, at some point, something might seem too much like what’s come before, and people will desire something new.

About the Author

Tom Vanderbilt is an avid writer and journalist. He writes for several publication and is a contributing editor of Wired (U.K.), Outside, and Art forum. He has also authored the widely popular book Traffic: Why We Drive the Way We Do (What It Says About Us) and Survival City: Adventures Among the Ruins of Atomic America. He has been a visiting scholar at NYU’s Rudin Center for Transportation Policy and Management, a research fellow at the Canadian Centre for Architecture, a fellow at the Design Trust for Public Space, and a winner of the Warhol Foundation Arts Writers Grant, among other honors.