Listen Up, Computers: You Still Can’t Beat the Human Touch

Almost 30 years ago, Electronic Arts famously ran an advertisement that asked, “Can a Computer Make You Cry?” It was a thought-provoking ad, but as far as I’m concerned, the answer is no.


Background image copyright iunewind  

Computers can’t elicit the same type of emotional responses that another living, breathing person can, but they might get you partway there. In the center of the latest “computers versus humans” debate is the use of algorithms for recommendation engines — especially among Web-based businesses, products and services that we use every day. Algorithms help us do everything from choosing our next read to finding a new job, but they often fall short; they just can’t replace pure peer-to-peer recommendations, emotional connections or personal experience. As more and more businesses are discovering, the beauty is in the balance of using an algorithm to collect, collate, even filter (somewhat) and then add in that secret ingredient — the human touch — to create a recipe for success.

The Internet, Your Personal Shopper

First, let’s address how algorithms can work effectively on their own.

Take e-commerce giant Amazon, for example. It doesn’t need human curation; its collaborative, filtering-based algorithm recommends toothpaste when you add toothbrushes to your cart and suggests “Lord of the Rings” if you’re looking at the “Game of Thrones” DVD. That’s enough for Amazon because its business is built on the fact that you can get anything and everything there. Amazon is the Web’s superstore, and as e-commerce matures as an industry, other online retailers are realizing that “there can be only one.” If they want to stand out and differentiate, they’ll need the human touch.

As a recent example, celebrity stylist Rachel Zoe just inked a deal with fashion site Piperlime (owned by Gap, Inc.) to hand-pick her favorite shoes, bags and clothing — recommendations that will surely be laid over the top of Piperlime’s engine to further tailor results to shoppers’ exact needs. Other cases abound where stylists’ selections based on body type, favorite colors or brands augment algorithm-recommended products.

You + Me = Us

More than 10 years ago, online matchmaking sites started using basic algorithms with feature-based recommendations to pair people up. For example, if a woman listed her hometown as San Francisco and her profile mentioned dogs and wine, this basic algorithm would spit out hundreds of men with matching characteristics. didn’t think this system was sufficient. It wanted its algorithm to be more specific and tailored, so it brought in psychologists and relationship experts who found that actions speak louder than words (in user profiles). Many users claimed to want one thing in a partner, but perused the profiles of complete opposites. That’s when started incorporating the human factor into its formula.

Now, when a man claims to only like blonde women between the ages of 25-30, will throw a mixture of looks and ages into its recommendations. According to’s success rates, it seems to be working quite well.

Feeling the Beat

On the flip side, let’s look at an industry that was traditionally built upon human curation, peer-to-peer recommendations and emotion — but is now getting infiltrated with algorithms: Music.

As streaming music services put nearly every song ever recorded within the tap of a finger, millions of listeners are depending on the services’ algorithms to help them navigate this endless sea of emotion-laden content. These algorithms analyze what makes two songs structurally similar — tempo, genre, etc., and identify sound-alikes, but once the algorithms have demonstrated their stuff, music fans are often left craving something more. They want deeper experiences, emotional context and something that moves them. Perhaps most of all, listeners don’t want to be bored; they want to be pleasantly surprised.

Back in the day, I worked in terrestrial radio at an “alternative rock” station known for breaking the rules. We surprised listeners by taking musical U-turns and dropping James Brown in the middle of our usual mix of Nirvana and Red Hot Chili Peppers. It helped us break through the clutter — and it worked. Listeners tuned in longer. They developed relationships with our DJs. They became loyal brand evangelists for us because we made them feel something.

Algorithms are essentially sequences of finely-tuned rules for organizing 1’s and 0’s. Humans, unlike computers, can break these rules. This is important because “breaking the rules” is how great music has always been made and how great listening experiences are created. It’s all about creating the unexpected.

An algorithm can do a great job of delivering songs, products, jobs or partners that match up with elements that you have search for previously. But don’t allow this to fool you into thinking that music service “robots” can identify why a particular song can make you turn up the volume, or get up and dance, or take a moment to reminisce. The myriad recommendation algorithms across industries spanning e-commerce, online dating and job hunting won’t create the kind of personal, emotional connections with customers that matter unless, of course, they add the human touch.

Humans (still) rule.

Kevin Stapleford is the Senior Director of Programming and Content Development at Slacker. Previously he has served as the VP of Programming for various terrestrial radio companies, overseeing such stations as 91X/San Diego and KNDD (The End)/Seattle, and as consultant for radio outlets, record labels and artist managers.

Must-Reads from other Websites

Panos Mourdoukoutas

Why Apple Should Buy China’s Xiaomi

Paul Graham

What I Didn’t Say

Benjamin Bratton

We Need to Talk About TED

Mat Honan

I, Glasshole: My Year With Google Glass

Chris Ware

All Together Now

Corey S. Powell and Laurie Gwen Shapiro

The Sculpture on the Moon

About Voices

Along with original content and posts from across the Dow Jones network, this section of AllThingsD includes Must-Reads From Other Websites — pieces we’ve read, discussions we’ve followed, stuff we like. Six posts from external sites are included here each weekday, but we only run the headlines. We link to the original sites for the rest. These posts are explicitly labeled, so it’s clear that the content comes from other websites, and for clarity’s sake, all outside posts run against a pink background.

We also solicit original full-length posts and accept some unsolicited submissions.

Read more »