With over 1,500 analytic professionals on staff, Chicago-based Mu Sigma boasts what has to be one of the largest collection of Data Scientists under one roof. Putting together and maintaining such a large group is no small task, considering the worldwide dearth of (and high demand for) Data Scientists.
But Mu Sigma got a bit of a head start on the competition. The company was founded back in 2004, at least four years before the term Big Data came into vogue and, with it, the role of the Data Scientist. Back then, companies were already struggling to derive actionable insights from growing data volumes, according to Mu Sigma Founder and CEO Dhiraj Rajaram. They just didn’t have the handy shorthand nomenclature to describe it as we do now.
Rajaram decided to leave his job as a management consultant at Booz Allen Hamilton and found his own firm dedicated to helping enterprises make better use of data in 2004 (side note: check out this video of Rajaram discussing the early days at Mu Sigma and his take on entrepreneurship). Fast-forward eight years and today Mu Sigma is, by Rajaram’s estimate anyway, “the largest pure-play provider of decision sciences and analytics” services on the planet. It works with over 50 Fortune 500 clients and last December closed a Series B round of over $100 million in new funding led by General Atlantic and Sequoia Capital.
As Rajaram described it to me in a phone call last week from his office in Bangalore, Mu Sigma provides professional and technical services to help customers to first identify potential Big Data analytics use cases, then build the systems and processes needed to exploit them, and finally execute and repeat the actual analytics. He said the company is “technology” agnostic, meaning it will work with its customers’ Big Data tools of choice, be it NoSQL approaches or more traditional data warehouse/business intelligence tools (though Mu Sigma consultants will recommend new tools customers should consider on a case-by-case basis.)
From a manpower perspective, Mu Sigma takes a hybrid approach. In most customer scenarios, a small group of Mu Sigma’s 1,500+ Data Scientists will join the client’s internal analytics team on-site, while a larger group will work offsite in Mu Sigma’s Indian office. Rajaram said this approach allows Mu Sigma to both gain a deep understanding of its clients’ challenges, their work culture and internal processes, as well as provide services as efficiently as possible.
I asked Rajaram about how long it takes for a typical customer to start seeing tangible results from working with Mu Sigma. He said it depends on the client and the complexity of the Big Data challenge at hand, but that even so Mu Sigma’s goal is to establish long-term relationships with its client. In other words, Mu Sigma is not the company CIOs should call in if they’re looking for someone to come in for a short-term engagement with the goal of training internal staff so they can operate they’re Big Data practice self-sufficiently. Rather, Mu Sigma is for those companies looking for strategic Big Data partners to work closely with building Big Data processes over the long haul. Mu Sigma isn’t the love ‘em and leave ‘em type.
There are pros and cons to Mu Sigma’s approach.
The Mu Sigma approach obviously benefits Mu Sigma in the form of long term revenue streams from long term clients, but it also benefits clients in the form of relatively fast Big Data time-to-value. In the majority of cases, Mu Sigma Data Scientists and consultants can help clients start deriving actionable insights from Big Data in a fraction of the time it would take to build an internal Big Data practice. They’re expertise should also dramatically cut down on the number of false starts that can doom Big Data projects from the get-go.
And the longer Mu Sigma works with a client, the better it should become at streamlining the process of identifying and developing the most valuable Big Data projects for that particular client. Indeed, one of the challenges that makes Big Data so difficult to perfect is the need for talented staff that understand technology and have deep business/domain expertise. Working side-by-side in the trenches with clients over months and years gives Mu Sigma Data Scientists a view into its clients’ businesses that most Big Data vendors lack.
On the other hand, as the saying goes, ‘Give a man a fish and you feed him for a day. Teach a man to fish and you feed him for a lifetime.’ The Mu Sigma approach has the potential to make clients dependent on a third-party for what many enterprises consider one of the keys to their competitive differentiation.
The question for CIOs, then, is do I want to (more-or-less) permanently outsource the core of my Big Data practice to a third-party or build up our internal Big Data capabilities so that we can eventually support ourselves? The answer depends on how quickly you want to start capitalizing on Big Data, whether or not you believe building an internal team of talented Data Scientists is a fool’s errand, and how comfortable you are exposing a third-party to some of your most valuable assets – your data.