Anindya Ghose, Stern Professor of Information, Operations and Management Sciences, Will Co-Chair AIG-NYU Collaborative Research Initiative

Anindya Ghose, NYU Professor of Information, Operations and Management Sciences and Professor of Marketing, is co-Director of the Center for Business Analytics, a Daniel P. Paduano Fellow of Business Ethics, and the Robert L. & Dale Atkins Rosen Faculty Fellow, all at NYU’s Stern School of Business.

Additionally, Dr. Ghose is one of two co-chairs of the recently-created AIG-NYU Partnership on Innovation for Global Resilience. He represents NYU, while Dr. Siddhartha Dalal is the co-chair from AIG. Both lead the Partnership’s Advisory Committee, which will select for funding those projects proposed by NYU faculty which demonstrate the potential to have a transformational impact on the world at large. Funded by AIG, the $5.5 million initiative will award $1.1 million in grants annually for five years.

Dr. Ghose’s research, which analyzes the economic consequences of the Internet on industries and markets, has received numerous awards and grants, including the National Science Foundation CAREER Award in 2007, a $2.12 million NSF grant in 2009, a $2.9 million grant from the NSF’s IGERT program in 2010 and five awards from Google and Microsoft.

Before joining NYU Stern, Dr. Ghose held positions at GlaxoSmithKline, Hewlett-Packard and IBM. He received his doctoral and master’s degrees in Information Systems from Carnegie Mellon University, an M.B.A. in Finance, Marketing & Systems from the Indian Institute of Management, Calcutta, and a B. Tech (with honors) in Electronics & Instrumentation Engineering from the Regional Engineering College (NIT), Jalandhar, India.

Professor Ghose recently met with reporter ML Ball to discuss his current research, the AIG-NYU Partnership and his love of the Internet.

What brought you from India to America, and when?

I came here in the year 2000. Before that, I was working for IBM in a consulting group, and before that, I was with HCL-Hewlett-Packard. So I’ve always loved IT.

Coming out of business school after my M.B.A., I wanted a job in IT, not so much in programming but in consulting or analytics, that sort of space. I enjoyed the first couple of years in industry but then I started exploring other professions where I would have the independence to choose my projects, have flexibility with respect to time, and be able to work on deep intellectual questions that may have bigger ramifications than what I was doing at the time. So all of that pointed to academics. To be in a good university in the US, you have to go through the rigors of a Ph.D. program, so I got my doctorate from Carnegie Mellon in 2000 and then came here to NYU in 2004, ten years ago.

What made you choose NYU?

NYU’s reputation is stellar, especially in the business school space, and in the two areas I work in, marketing and IT, it is pretty much considered at the very top, if not the best. And New York was a big draw. I grew up in big cities for the most part, Bombay and Delhi and Calcutta. Plus, I knew some of the people. The synergy was quite explicit and I realized this was the place where I would be able to be most successful.

What interests you most about data science?

When I started doing my Ph.D., the one area that really fascinated me, and still fascinates me to no end, is the Internet. I have always been totally enamored by new phenomena that are cropping up on the Internet because of its underlying infrastructure: new business models, new forms of communication, new forms of collaboration. And every new phenomenon brings with it a swath of fascinating new data.

I’ve always tried to look for the next frontier. Today, “topic X” is the most cutting-edge recent phenomena, so I want to work on that. But I also want to work on, and anticipate, what would be the most cutting-edge thing in the next two years. I need to start working on that today because getting access to data and negotiating data with companies takes time, and if I wait until everybody recognizes something as “hot,” I might be behind the curve.

Have you been able to accurately anticipate trends?

I would have to say yes. A lot of people can see what’s coming. But it’s not just important to see it coming, you also should find it interesting. And somehow or other, I have not only been able to predict what was coming but also found it interesting enough to work on it, to collaborate with companies to get access to data and run field experiments with them.

What have your research predictions led to?

For business school academics, the single most important criteria in research accomplishments are journal publications. Specifically, premier journal publications that are rated top tier by every university in the world. So that’s our target. We also present at conferences, but those take a second seat.

Second in importance would be recognition, in the form of prestigious awards or grants, from NSF and the corporate world. My research has been funded extensively by Google and Microsoft, as well as by many other companies in industry.

I think what really motivates me increasingly these days is to work on a problem that is very real, where I can take the findings from my research and apply them in the companies that gave me the data in the first place.

Have your research results been applied to industry, affecting business outcomes?

Yes, that has happened extensively in the last four to five years. My work takes me overseas a lot, particularly to China and South Korea. My most recent work has been in the mobile computing space, in which China and Korea tend to be like crystal balls for the rest of the world. What happens over there today will happen in the U.S. about a year or two from now, and in Europe, a couple of years from now.

One of the most recent projects I’ve done where the research results were then directly applied was exploring only-channel symmetries in the world of digital advertising in South Korea. We were studying whether, when people get exposed to an advertisement by a brand in one channel, then also get exposed to the same ad or the same brand in a different channel, that increases the propensity to buy the product. Or does it actually reduce it because of the annoyance effect?

We learned that if people are shown an ad in a different format in a different channel, even though it’s the same message, our brains tend to process it differently, so we don’t face that wear-out effect. Rather, there is a reinforcement effect in our mind which actually increases our propensity to buy. This is something that companies in South Korea asked me to investigate. They wanted to know if there are synergies in the first place, and if so, can they be measured. So we ran a number of randomized field experiments to causally ascertain the synergies. We then shared our findings with the companies, which got very excited. They executed our study themselves and saw similar results.

Another study I did recently was with companies in Germany which were measuring the effectiveness of mobile coupons. You can now target people with real-time coupons based on their location. For instance, as I’m walking past the shopping mall with an iPhone, stores can sense that I’m walking past them and they can send me a coupon. Some firms are beginning to do this in the US; they are not yet as sophisticated as the folks in Korea or China but it’s only a matter of time.

In the German study, we looked at the effect on location on smartphone coupon redemption. We then went back to the companies and told them how they should run their experiments and change their prices. They did, and saw a lift in their redemption rates. So you can meaningfully design studies to not only get good research out of them but to also have an impact on industry.

Can you describe the AIG-NYU Partnership and your role in it.

I am the Chair from NYU, and Siddartha Dalal is the Chair from AIG. Obviously, AIG has been a leader in the insurance space for a number of years, but recently, they have set up a data science team. In the process, they have become very excited about collaborating with academics, especially asking top universities to help their own science team jointly figure out answers to problems of interest to them.

As they talked with NYU, they started plotting the idea of joint research collaborations on transformative projects that would be of interest to professors at NYU and the data science team at AIG. During those discussions, we determined that this would be a long term, five-year collaboration.

The initial discussions took place with Dr. Paul Horn, Senior Vice Provost for Research here at NYU. He really championed the whole thing and formed the steering committee across NYU disciplines. He also chose me to be the Chair from NYU; I’m very flattered and honored that I was chosen. I’m excited to work with a stellar group of advisory committee members, the Who’s Who of NYU. On the AIG-NYU team, we have nine people from NYU and a similar number from AIG. It’s been very enlightening to figure out how we can actually work on problems of direct relevance and applicability with a company like AIG. Right now, we’re in the process of requesting proposals, from NYU faculty but also advisory members. Soon, we will meet together and look at the proposals which have been submitted, and then hopefully select the majority of them.

I anticipate there will be a strong response to our call for proposals; the deadline is April 11th. There is already so much buzz about this, as it’s being circulated widely across NYU, and I would love to see a lot of people being funded.

The AIG funding is for five years. Will some of the projects last longer?

Yes, certainly. Some projects might show meaningful results in two years, but others might take three to five years, some maybe even more. So this is a long-term process. I won’t be surprised if after five years, AIG is happy with what we have accomplished and asks us to renew for another five years.

My plan is, after a few months, to organize a workshop/symposium, inviting those faculty whose proposals have been selected to present their most current state of research that has come out of the funding. We would probably do that twice a year; research projects take a long time, and twice a year would give us a sense of where they are in the early stages and where, subsequently, they will be a year from now.

The second step involves AIG figuring out which of those funded projects can be applied directly in their companies. Can they be monetized? Can the ideas be productized? Can they incorporate them into their current data science initiative?

What’s the benefit of the partnership to NYU?

This program is unprecedented. NYU has never had this sort of project, of this scale and scope: five million dollars over five years. It’s a big deal. Imagine the number of faculty whose research can be funded out of this. The National Science Foundation has dramatically cut down on funding for faculty, meaning that federal research funds are slowing down considerably.

Because of this, there are many, many faculty to whom even a small-size grant would make a huge difference for the potential of completing their research. That is why I personally would like to make this as inclusive as possible, rather than making it overly selective.

In essence, the fascinating research you can do with this data is limited only by your imagination. For us geeks, this is amazing. For me, it applies directly to the new science of cities, which is something I’m currently working on. I am very interested in figuring out how people in cities live. When do they wake up? When are they the most energetic? When do they party? We can now measure this because of available technologies, like smartphones. Every time your phone is with you, it’s streaming out a ton of data about where you are, what you’re doing. So by combining that with available technology data, you can literally map out and visualize how a city breathes, when it is most alive. That’s the kind of research I’d like to be doing in the next few years.

 

By ML Ball