Social networks and engineering serendipity in the workplace


The New York Times has an interesting article titled Engineering Serendipity which looks at the some of the ways companies are trying to create felicitous and unexpected connections between their staff. After introducing what Yahoo! and Google are doing in the space, the article continues:

As Yahoo and Google see it, serendipity is largely a byproduct of social networks. Close-knit teams do well at tackling the challenges in front of them, but lack the connections to spot complementary ideas elsewhere in the company. The University of Chicago sociologist Ronald S. Burt calls these organizational gaps “structural holes.” In a 2004 study of 673 managers at the defense contractor Raytheon, Mr. Burt found that managers who serendipitously bridged such gaps were more likely to generate good ideas (and advance professionally as a result). “This is not creativity born of genius,” he wrote. “It is creativity as an import-export business.” In such cases, serendipity is the spontaneous plugging of these holes, over which good ideas flow.

The article describes some of the research being done in the space by measuring online and real-world interactions:

One discovery, says Ben Waber, a co-founder of the company and a visiting scientist at M.I.T., was that employees who ate at cafeteria tables designed for 12 were more productive than those at tables for four, thanks to more chance conversations and larger social networks. That, along with things like companywide lunch hours and the cafes Google is so fond of, can boost individual productivity by as much as 25 percent.

“If you just think of serendipity as an interaction with an unintended outcome, you can orchestrate pleasant surprises,” says Scott Doorley, a creative director at Stanford University’s Institute of Design. He and his colleague Scott Witthoft have instituted simple measures like positioning couches near doorways and stocking rooms with multiple types of seating to encourage lingering conversations.

Serendipity has long been a key interest of mine. In a Living Networks Forum I ran in New York in 2003 we created “enhanced serendipity”, and I later wrote in more detail about Creating enhanced serendipity.

As brought out in the New York Times article, this relates strongly to the domain of social networks, which I have been studying since the 1990s. Arie Goldshlager, who pointed me to the New York Times article, asked me to respond to a question he asked on Quora, Can you engineer serendipity?

My answer on Quora is reproduced below:

Yes you can ‘engineer’ serendipity.

Serendipitous connections result in happy and unexpected outcomes. This requires the design of a network structure in which nodes (e.g. people) that have a higher likelihood of connecting felicitously are exposed to each other. In a simple sense a greater network density (e.g. smaller workplace) results in more connections so is more likely to give serendipitous outcomes. However this also increases unproductive connections, which can create a real drag on productivity. In addition, increasing connections between a given pool of relatively heterogeneous people (e.g. employees) is far less likely to create serendipitous connections than enabling connections with a constantly changing array of external nodes.

As such engineering serendipity can be done to a limited degree through workplace design, however the real magic happens when you engineer consistently diverse exposure across a given pool of people, and also bring in diverse external connections.

A key issue is the degree of similarity of nodes (e.g. people) across a range of dimensions, including aspects of personality, interests, and work experience. True engineering of serendipity comes when you experiment with connecting nodes with differing degrees of similarity to find what creates the best outcomes in a particular context. This requires individual profiling, and also tracking the results relative the degree of similarity of people who have connected.

There is much fruitful work to be done in better engineering serendipity.

We now have deep accumulated insights from the field of social networks analysis, combined with ever-richer social data and a pressing drive for companies to drive higher productivity. We are now at a point at which the engineering of serendipity in the workplace and beyond is attracting substantive efforts and some promising early outcomes. It’s an exciting space which will develop far more in coming years.