Kathryn Brohman: Finding Value in Open Data
Kathryn Brohman is Associate Professor and Distinguished Faculty Fellow of Management Information Systems at Queen’s School of Business. Brohman has built an impressive research record in the areas of data warehouse usage and data strategy, net-based customer service systems, and project management. She is currently working on one project with Queen’s colleagues that looks at how “elastic services” can be enabled in cloud computing; it has attracted three-year funding from the Natural Sciences and Engineering Research Council of Canada and five-year funding from the Ontario Research Fund.
In this conversation with QSB Insight, Brohman discusses the challenges relating to how, in a mobile and cloud computing world, consumer information is collected and converted into new offerings.
On rethinking the nature of data ownership
Companies have been collecting data from consumers from any source they can find and now these organizations are sitting on terabytes of datasets and not really knowing what to do with them. They’re saying, We need to analyze it, figure out ways to get data scientists to mine it. To me, that’s nothing new. That’s been around forever; banks I worked with 10 years ago had a group of data scientists doing that. But the idea of actually putting their data out for everyone to see is very foreign. They see it as a resource and, like all good organizations, they protect their resource.
For the last five years, my big question has been, Why are you doing that? Generating value from data comes from actually connecting, getting value through the nodes, getting that data to the right person at the right time with the right message.
The idea of open data is new to a lot of the private sector industries but seen as a huge opportunity for new ventures. There was a company I worked with called Tripit.com. They created an entire business on integrating travel data for consumers. As a consumer, your travel arrangements can include airline, hotel, and rental car, and you get three different reservations and confirmations. Tripit said, Whenever you get one of those email confirmations, send it to us. They parse all your emails and put them into this structured database form. When you call it up on your mobile web service or whatever you’re using, Tripit gives you an integrated view of all your travel information.
They’re co-creating value with me as a consumer with data that they don’t even own. This is becoming more and more common with start-up companies. Once Delta sends me that confirmation, isn’t that my data? Delta’s the one that actually incurred the cost of creating the data in that form, so the processing costs of actually putting the data into the email was their cost. But as soon as they sent it to me and then I forwarded it to Tripit, who owns the data?
They’re co-creating value with me as a consumer with data that they don’t even own. This is becoming more and more common with start-up companies.
This got me into questioning this whole idea of data ownership. Maybe 10 years ago, the Bank of Montreal was considering what they called screen scraping. If I, as a consumer, were to give them my Royal Bank account login and password and my CIBC account login and password, theoretically their systems would use my logins and passwords, scrape that information, and give me one profile view of all my financial information. The data people saw it as valuable but the business didn’t. They thought, Well, if we’re not going to get paid for that, why on Earth would we do it? Now you see companies like Tripit and financial companies providing that service and one financial view.
On building a really smart dashboard for life
What I’m finding even more interesting now are these web services that integrate data and make things more convenient. You get what I call data completeness, a complete view of whether it’s things in your closet, or your medical information — all the different prescriptions that you’re taking — or your nutrition information. These are all integrated data platforms.
It’s all great and well to provide an integrated view of someone’s financial information. The harder problem is actually making the service elastic in terms of adding value for the consumer. An example is loyalty cards in the hospitality industry. I have loyalty cards for a number of different hotels. With a web service— or what I call an elastic service — I could prioritize these loyalty cards in terms of my own personal preferences as well as list other preferences that guide my hotel reservation choice. The application could review the hotel websites for offers, identify a short list of offers that best meet my preferences, and manage the booking of the hotel I choose. It depends on my own personal hotel reservation strategy, but my vision is that these elastic services will not only show you the information but actually do transactions for you on the fly.
There’s a whole lot of data available but there’s not a whole lot of services available that actually do anything of value for people.
Once you go on the consumer end, you’re asking competitors to talk about their information being shared, and they don’t like that.
On where innovation will likely happen
A lot of times companies don’t want [data sharing] to happen. With the consumer having all of that control, it challenges traditional organizational boundaries and capabilities. That’s why we’re seeing service innovation happen in these new start-up companies.
The governance issue around sharing of data is a lot less challenging for organizations when they think about these types of elastic services internally. But once you go on the consumer end, you’re crossing organizational boundaries. You’re asking competitors to actually talk about the idea that their information is being shared, and they don’t like that.
The one industry that I could see doing it would be retail. One of the big Canadian companies is trying to become the data provider for all retail companies in Canada. In the retail space, they see each other as competitors but they’re kind of complementary. If I’m going to fly to Florida, I’m either going to fly on American or Delta. Whereas in the retail space, if I’m going to wear a coat from Michael Kors, I may also wear a coat from Hugo Boss.
Any industries that are going to lead an initiative like this are industries that offer complimentary products where an individual’s service or product portfolio is made up of services and products from multiple companies.
On the key issues in data management
I think the cloud’s done. People are still implementing cloud capabilities but conceptually they get it. The pressing issue now is Big Data, which is a term I don’t like. The idea of data ownership and governance and who owns this responsibility in a company and why do we have all this data and doing nothing with it. Those are the big questions. I would say we’re three or four years away from the organizational boundary question. Even chief data officers are still talking about their own firm’s data. When I ask them, What about other data?, they answer, You mean like Twitter data? No, your competitor’s data. They’re still so constrained by the organizational boundary.
I remember being in a meeting with a Delta executive when I was at the University of Georgia, this would have been around 2000. I asked these executives what they thought the definition of customer services was in the travel industry. They talked about comfortable seats and access to food and peanuts, and they asked me, What’s your vision of customer service? I said it was walking into an airport without a ticket and getting on the first flight available for a competitive price. The whole concept of walking in with your phone, your phone knowing where you want to go, the airport knowing all the flights that are going to your destination, and to be able to book on the fly. Give me three different options, boom, I’m there. The whole concept of sitting at a Delta gate and watching an Air Canada flight fly to Toronto is ridiculous.
We think there’s no way any of the regulated industries are ever going to share information like that. But from a technology perspective, we can do it. I’d love to see it in my lifetime, but I don’t know.
— Interviewed by Alan Morantz