A month or so ago, with little fanfare, Mu Sigma, a data analytics company, reached the $100 million high water mark in revenues. Around the same time, it also became a ubiquitous presence in the US, sponsoring and speaking at several important industry conferences such as Enterprise 2.0.
Seemingly out of nowhere, this Northbrook, Illinois-based company that provides decision science and analytics solutions, with a large offshoring centre in Bangalore, has become not just one of the hottest young technology companies around — its success has also become a beacon for what the next big wave in outsourcing will be.
In general, arts and science never see eye to eye. The debate on whether Leonardo da Vinci was a scientist or an artist often raises temperatures. However, when Dhiraj Rajaram started Mu Sigma seven years ago, he was not just creating a company that would provide decision science, but also one that would have both arts and science capabilities.
Mu Sigma is just that, says Rajaram; in addition, it has scale. Art, according to Rajaram, represents design capability and science represents the laboratory. Scale is the ability to work like a factory. “All this is coming together for the first time. Not only in India but anywhere else in the world,” adds Rajaram.
Every day, we create 2.5 quintillion bytes of data and 90 per cent in the world today has been created in the last two years alone. This data comes from everywhere: Sensors used to gather climate information, posts on social media sites, digital pictures and videos, purchase transaction records, and cell phone GPS signals, to name a few.
The problem is, unstructured data — text, videos, tweets and social media — has grown at phenomenal rates. Google and Facebook, for example, are the biggest repositories of consumer behaviour data. If companies can find valuable insights within this seemingly unintelligible torrents of bits and bytes, they will be able to make more informed business decisions, whether in marketing, sales or practically any functionality within a company. This is what Mu Sigma does in its back office in Bangalore: It sifts through mountains of information and analyses it for clients.
“People are looking to make better decisions. The amount of data that is required to create this decision is exploding every day. To make this happen, one needs to have an inter-disciplinary talent of mathematics, business and technology. Every player wants to be able to do this but they do not see all these three things together,” he says.
For instance, Mu Sigma helped a New York-based online clothing store spot shopping trends among its customers that in turn radically influenced its website architecture and merchandise selection. Its data analysis skills have already attracted a range of clients, from Dell to large insurance companies to Microsoft’s Bing division.
Facilitating more sound business decisions is, not surprisingly, good business. According to a recent IDC research, business analytics grew at a compound rate of 9.8 per cent per year, hitting $35.1 billion this year and projected to reach $50.7 billion by 2016. For 2011, analytics underwent a 14.1 per cent increase in revenue. Mu Sigma impressed enough to attract a total of $150 million in private equity so far — in December last year alone, General Atlantic and existing investor Sequoia Capital announced that they were pumping over $100 million into Mu, while Accel, an early investor exited its stake.
Acting on a hunch
While it may seem that Rajaram has hit the jackpot, he says that when he started Mu’s business plan (2003-04), data was not as big it is now. “It was a hunch that this is where the world will be. And in the world of tomorrow, knowledge and learning will be far more important than knowing,” he adds.
Rajaram, a engineering graduate from Anna University, Chennai, and an MBA from the University of Chicago, blames his decision to quit a comfortable job at Booz & Co for what he calls ‘middle-aged men menopause syndrome.’ Quitting his job was not enough. He sold his home in Illinois to raise initial funding. “I realised that calling people and telling them that you want their business and what can you do for them is the easiest way to get work. I did not have any great contacts in the industry, was just 29 and deployed all my savings of around $2.5 million into this company. We were talking to 20-30 prospective clients. And Microsoft happened to say ‘yes’ first,” he reminiscences.
Back to school
Since scale is a critical part of his business plan, India was an integral part, too. “India is not an after-thought. India is a fore-thought for us. This company could not have been built without India. You need scale and you need to train a lot of people in arts and science. People who understand English but are not afraid of Mathematics,” explains Rajaram.
Scaling up has meant hiring employees — Mu Sigma has around 1,700 of them. According to a recent McKinsey study on Big Data, India will require nearly 100,000 data scientists in the next couple of years. Rajaram realised early on that he would have to invest in creating a talent base. Doing so entailed methods that were a little out of the ordinary, to say the least.
A candidate applying to Mu Sigma needs to clear four stages of recruitment and a training program that consists of three parts. Under recruitment, the first stage consists of an aptitude test that has basic arithmetic and logical problems. This is followed by a group discussion, “but our group discussion aims to get people out of their comfort zone. While we want people do to the maths, we also want them to be able to talk,” adds Rajaram. The third stage has a video synthesis, where a candidate is shown a video and then provided a piece of paper slightly bigger than a visiting card. On it, the candidate has to synthesis whatever he or she has just seen. The fourth is the interview, which the company calls a ‘Fit interview’. Of the 100 candidates who apply, only a handful get selected.
Then comes the next big hurdle: passing what the company calls Mu Sigma University, This has three parts. First, the candidate is introduced to consulting principles. The curriculum of the second part consists of analytics, which covers areas like artificial intelligence, business intelligence platforms of SAP and Cognos. Finally, a mini MBA course has to be negotiated. “The entire process takes two to three months. This year, around 1,000 people will pass through the Mu Sigma University. We are creating a new breed of skill sets in the industry. We are changing the world, just like TCS and Infy did in 1990,” says Rajaram.
Whether it can become the next Infosys depends on how successfully one can sustain this model, which Rajaram says is not just about cost arbitrage. “What we are doing cannot be even termed as outsourcing. That is why it is interesting, as it is a category-defining company,” he said.
Not everyone agrees. In an online article titled ‘Can Big Data Be Outsourced? Mu Sigma’s $150 Million in VC Backing,’ Peter Skomoroch, a principal research scientist at LinkedIn, focusing on building data-driven products, as well as a founder of Data Wrangling, which offers consulting services for data mining and predictive analytics, says, "I'm sceptical of the idea of end to end 'analytics outsourcing' right now.” Skomoroch feels that vision and creativity are unlikely to be commoditised any time soon. “The competitive advantage in this latitude will go to companies that establish unique data sets and build teams that are aware of how to leverage them. The most plot-changing analytics is going to happen from a small set of talented individuals, not an army of contractors,” he adds.
This won’t deter the ambitions of Rajaram, however, which are considerable. “I want a little bit of Mu Sigma in every company. Like Intel Inside,” he says.