Serendipity as engineered randomness

July 30, 2014

Last year Greg Lindsay wrote a wonderful piece  for the New York Times that described how a number of organizations  were seeking to ‘engineer serendipity’ - either through modifying their  physical workspaces, or by instituting processes that enrich for  fortuitous encounters.

Sound familiar? Of course it does! Conceptually this is one of the reasons for Lunch Roulette (in all it’s implementations).

Where am I going with this I hear you grunt? All of the  processes/tweaks described are great - and will most certainly enrich  the number of 'random’ interactions, and by extension serendipitous ones  - but it’s important to note that the sought after serendipity is a  second order effect.

What do I mean by this?

One of my favourite definitions of serendipity is the following: “the  occurrence and development of events by chance in a happy or beneficial  way”. Serendipity is a second order effect (the ’development’  here is key) because, it’s not enough to just be a participant in a  random act, you have to be aware, or made aware, of the act - and be  able to contextualize the act in a new and somewhat unexpected way. This  is even evident in the roots of the word. The Three Princes of Serendip  were latter day Holmesian observers, “always making discoveries, by  accidents and sagacity, of things they were not in quest of.”

This observation is important, I think, for a couple of reasons.

For organizations to engage in this, it has to be beyond "engineered  randomness’. There has to be a component devoted to increasing, and  maintaining, individual awareness and mindfulness. Only when you’re  truly aware of your surroundings can you really make any serious  progress in an unexpected direction.

Secondly, assuming a heightened level of awareness and subsequent  translation of random acts into serendipitous ones, the organization has  to have in place a measurement scheme to associate the ultimate outcome  to the conditions that contained the act in the first place. Without  this, it might be akin to random, engineered randomness and really, who  wants that?

How then to instantiate such a measurement scheme I hear you curse?  Good question. At the moment I am not sure, but maybe I’ll have some  thoughts in time for next weeks thrilling installment.

Thanks for listening,

DT