This Curation Trend has One Big Problem: Scale

By Erin Griffith , written on September 10, 2012

From The News Desk

Like it or not, the curators are coming. And I'm not talking about the kind that work with

The latest wave of curation goes beyond Drudge Report-style link aggregation. It's about adding some sort of value to the content you're curating, either by the choices you make, the method of delivery, or what you do with it once it's curated.

Before I go any further, let me just say: It would very easy to dismiss the new wave of curation. Like Smankers, the burgeoning curator population is filled with some grade-A morons hell-bent on giving everyone a bad name. In many cases, what they call "curating" is nothing more than decision making. I loved Choire Sacha's characterization of Web curation as "a precious bit of dressing-up what people choose to share on the Internet." Having a few well-maintained Pinterest boards does not make one a curator.

Indeed, at the Curator's Conference earlier this week in New York, the word got tossed around so much that it became comical.

I love the way you curated that outfit from your wardrobe.

Thanks. That was a really well-curated sentence.

I know. I curated the words from my vocabulary.

Genius. Let's curate EVERYTHING.

Totally. Hashtag #Cure-gasm!*

But let's dismiss those cases as exceptions. Look at the early days of Twitter and Facebook: We had no shortage of wacky social media gurus, ninjas, and wizards to parody. You know the types -- the old-time marketers with 120 followers on Twitter who put out a shingle as an expert on everything social, promising Internet fame and riches. Today those people are still easy targets, meanwhile other, less annoying people have built massive, valuable companies in the field. Chief Marketing Officers are poised to control a larger chunk of the budget than IT, and their top concern is social media. Fast Company declared yesterday that, according to a McKinsey study, social technologies can create $900 billion to $1.3 trillion in value for companies.

There is a parallel to be drawn here. Dismiss the early, annoying adopters of a trend, and you may be dismissing the next massive shift in media creation and consumption. Dismiss the annoying curators, but not the field of curation. 

Curation continues to gain momentum in the tech world. I've seen startup after startup assess their algorithmic recommendation engines and conclude that a human touch just works better. While Amazon, with its robust recommendation engine, still peddles suggestions based on past purchases and searches, other smaller companies have found that it takes a human to know what other humans find interesting.

Take Songza: the company tried user-generated playlists and Pandora-style algorithm-built playlists. But its service didn't take off until it employed a team of experts to curate its mega-popular activity-specific playlists. Sulia surfaces its news from experts -- it picks those experts based not on followers, or engagement data, or whatever black box Klout uses to measure influence, but from the one human-built indication of expertise on Twitter: Twitter lists. Apparel trade show Capsule has become an exclusive fashion event of the year because of its limited, curated group of presenters. Brand new startup Fun Org began with algorithm-driven activity recommendations for people but quickly switched to a select set of offerings gathered by humans. Behance uses a team of full-time curators to choose which creative work to feature on its homepage. YouTube's homepage is no different. And on and on.

The problem these tech companies -- and their investors -- struggle with is that curation requires humans. Humans are expensive, and they don't scale. The whole promise of Google was that IT could use data to automate just about anything. As tech companies begin to look beyond the algorithm, they're grappling with the problem of scale and human capital. Computers can't always do the job. That was the promise of crowdsourcing -- everyone from news organizations to online communities like Digg turned to the wisdom of the crowds to automate work once done by a human. Now they're learning that's not always the best solution.

Behance faces that issue. CEO Scott Belsky said Behance's front page could no longer display what algorithms determined was the most popular art within his site's community. Because of boobs. They are universally the most popular thing on the Web, and not even a tasteful, creative site like Behance is safe when the "wisdom of the crowd" is involved. To be clear -- boobs are welcome on Behance, but the site skews toward commercially viable work. A porn pit may entice creative directors but not in the way Behance wants to entice them.

"There is a real flaw in leaving curation to the masses," Belsky said at the Curator's Conference last week. "Giving experts more weight can bring better things to the surface." So he hired a small team of design and art experts to and curate the site's best content for the front page. Their jobs have become increasingly meaningful as Behance grows: Getting featured on the front page can change an up-and-coming artist's life, Belsky said. But with that growth comes challenges -- how can a small team sort through a massive library of content that's growing at light speed?

The company's solution is a human-tech hybrid. Behance built a system of queues to help its curators spend as much time as possible looking at images, rather than actively navigating the site's front-end for content. "The tech isn't an end for the company. It's a tool to help with mission," Belsky says.

Capsule, which organizes a sort of trendy curated flea market, but for apparel wholesalers, is not a tech company but faces the same issues of growing while curated. "We'll never be bigger than we are now, because that would kill the reason for our success," Director Chris Corrado says. "People need to know we are a line in the sand." Capsule will consider launching new verticals to grow but will not expand its existing trade shows.

A business based on human curation will always be smaller by definition than one that scales through algorithms and computers. But that human element makes what they're building that much more valuable. High end luxury products companies like BMW, Chanel, and Rolex will never be as big as Ford, H&M, and Casio.  But that's fine, indeed desirable. Because tech can be replicated. Taste cannot.

[*Conversation may have been slightly embellished.]

[Illustration by Hallie Bateman]