Big data, characterized by the dramatic growth in the volume of data (internally generated and from external sources) available to businesses, presents new opportunities for banks to grow revenues through better customer insight. But can banks create the corporate cultures necessary to compete in the big data era? Bank Systems & Technology editorial director Katherine Burger spoke with Allen L. Weinberg, a New York-based director in the business technology office of McKinsey & Co. and co-lead of the consultancy's banking and securities operations and technology practice, about what it takes to compete successfully on data.
How do you define "big data"?
Weinberg: Big data is the notion that in a more digital world, where people are using mobile devices, interacting with you online, and where you have the capability to analyze live call center recordings to pull the data out -- it's the collection of all of that data from a variety of sources, [which] could be internal and external. [And then] pulling that together, using much cheaper and rapidly improving tools, to get to real customer insight, and to drive better decision making and performance. It's not just about what I can do for the customer, but [also] internally within the bank: How do we make everyone more productive? How do we drive the effectiveness of every decision we make?
Recognizing the value in data and trying to do more with their data is not a new concept for banks. What is different today? Is it improvement in the technology, the external pressures of competition and regulation, or something else?
Weinberg: It is a bit of all of the above. It is all of those things coming together and adding the notion of the data-driven culture. So we point to Amazon, Tesco's or Harrah's -- examples from other industries. They basically take every decision they make and try to, where possible, inform that by data, and be open to where the data comes from, whether internal or external.
What's interesting for banks -- with the regulatory overlay, the additional complexity, the challenges to profitability [and] the economic outlook -- [are the opportunities to] drive growth and add a customer. Now that all customers to some extent have a capital cost associated with them, who I'm adding is incredibly important. So where it really starts to make a difference is, if I can be smarter about who that customer is that I'm adding, how do I get the most cross-sell from that customer? That's a competitive differentiator. The small differences will be much more magnified because of the capital overlay, so the relative profitability of customers -- the relative ability to capture share of wallet -- is going to matter a lot more than in the past. The notion that I can really understand and estimate the value of a customer, based not just on my relationship but also on other outside data, starts to really matter.
In what other ways does big data change the competitive environment for banks?
Weinberg: The organizations that can mine transaction data to really understand behavior -- this changes everything. The threat is, first, that someone just gets much better at picking off your customers. The second is, done the wrong way, people spend a lot of money on this and do not get return on their investment. Third, which is harder to see now -- does a technology player come in and have interesting product offerings they bundle? That is a real possibility.Banks have often thought the internal data they have [provides an advantage]. What we're finding is, there are external sources of data that will be equally valuable. I don't necessarily need to have millions of customers and data on them to be able to target people. So suddenly what I thought was a proprietary asset to me is no longer proprietary; someone else can get it without the cost to build it up, and maybe with better tools to access it. That's a bit of a game changer.
What kinds of organizational and culture changes are required for banks to meet these competitive challenges?
Weinberg: Where this works well is a model in which everyone sits in the room together and is looking at the data and testing insights around consumers. It's got to be the data scientists sitting with the business folks. It's breaking down all those silos, [to have] a customer-based view, regardless of where the data happens to sit. It's structural -- not getting stuck in organizational silos, pulling some of these functions together. It doesn't have to be an organizational chart change, but it's at least getting these people talking together and sharing across the data. You [also] need a business champion who is going to drive it through.
There is new technology to bring in. There's a whole ecosystem of capabilities that are being developed. It doesn't have to all be developed in-house, but the internal IT folks are an important part of this.
Lastly -- it sounds simple but sometimes isn't -- where do you want to start? In most cases you don't need to have a big bang. In general it's most effective in banks where it's a pilot here or there, because it's an enormous amount of learning about how you do this. It's picking a spot and going after it.
What kinds of data security and privacy issues must banks address in the big-data era?
Weinberg: [Previously] our data was in many places, so if someone wanted to find it, it was hard. Now I'm bringing it together in one place, whether that's internally or externally in the cloud. It becomes a more valuable target, in terms of information that someone wants to go after. In some ways drawing the value out of the data has the problem of increasing the value of the data, which makes it more valuable as a target.
From a privacy point of view, you have to be a little bit careful about what information you are collecting about people, how you are using it, what decisions you are making. There is a whole set of compliance issues. There are some things people will give voluntarily, but then you need to be careful [about]: … How do I want to use that?