Discover PerformanceHP Software's community for IT leaders // March 2014
Culture check: Is your business prepared for Big Data analytics?
Next-generation analytics will tell you a lot about your business. Is your business prepared to understand, hear, and act on information?
The actual tools to help you extract meaningful value from your data have gotten steadily better. In 2014 and beyond, the abundance of technology is adding a level of complexity and confusion about finding the right tool for the job. This links directly to the need for people who have the skills to understand the technology and the scope of insights it can deliver.
Discover Performance recently spoke to Colin Mahony, vice president and general manager of HP Vertica, who shared his thoughts in our January issue on 2014 as the year analytics really goes big. This month, we talked to him about staffing a data science team and best practices to transform the business with real-time information.
Q: We hear frequently about the importance of the data scientist’s role in next-generation analytics, and yet there is an acute shortage of talent. Do you have any advice for executives who are having trouble filling their data science positions?
Colin Mahony: So far, we’ve been leaving a lot of the data analysis up to these data scientists with PhDs. It’s wonderful to have those people, but for Big Data analytics to go mainstream, the average person needs to be able to use it. And the average person doesn’t understand how to do those correlations, and shouldn’t need to learn.
People are inherently visual. If you visualize a correlation with a heat map, that effectively communicates the same information as mathematics concepts, such as P or R values, but in a way that the average person can understand. Business intelligence tools have come a long way in making visualization and interaction with data easier, so that you don’t have to be a data scientist to benefit and take business action. This is especially critical as the time between analysis and action shrinks to near real time.
Q: So the technology is bridging the human gap, in lieu of a lot more PhDs…
CM: The software is going to help scale, just like it always has. Software is what’s going to fill that gap between the people that have the technical knowledge and the rest of the world that needs to benefit from it.
I think every software application is going to have embedded analytics. And because of that, we’re not going to need all the data scientists that we just don’t have enough of.
Q: How should an enterprise approach a Big Data analytics project from an organizational perspective? What are the best practices you've seen? Does philosophy matter?
CM: To be successful, the culture of the company has to value analytics and the power of information. So, first, before you talk about any technology, ask yourself: "Does the corporate culture value the power of information and what it can do?"
If it does, the second step is to ask: "Is the organization able to take that information and get it to the right place at the right time to change the outcome of the business?" If your data project is siloed off in the corner with a data science team, often you’ll find very meaningful data that could move the business forward, but because it isn’t plugged into the business, timely decisions and actions don’t happen.
Q: So real business integration is the key. How do you make that happen?
CM: Our customer Zynga pioneered the most successful model I’ve seen. They embedded a data scientist—a statistics person—into every one of their business units. So during a meeting or a scrum, the general manager of the team is sitting next to a data scientist, who also sits amidst all the gaming developers, and this person was running real-time statistical modeling against all the user data. They would determine what experiences the gamers liked better, and they would literally feed that data back to those teams in real time so that the developers could start changing the games to fit the experiences that people want. This is why these games are so loved.
Insurance, as another example, has always been very siloed, where an actuary determines what the rates should be. More nimble companies are using more accurate data to calculate risk in near real time. Some of our clients offer a black box that goes in your car, which tracks your driving behavior, analyzes it, and rewards safe drivers with cheaper rates. At the same time, this data broadens the carrier’s understanding of driving behaviors, which influences policies and rates for specific demographics, etc.
Q: In other words, the challenges going forward have less to do with the technology of analytics, and more with the human and organizational aspects of making analytics pay off.
CM: The two most important factors from my perspective are, one, figure out how data can be a core part of the business and, two, structure the business so you can take advantage of that information at the speed and scale that will drive significant competitive advantage. The data shows that leaders who leverage data and act quickly win big.
Welcome to a new reality of split-second decisions and marketing by the numbers.
Looking toward the era when everyone — and everything — is connected.
Introduction to Enterprise 20/20
What will a successful enterprise look like in the future?
Challenges and opportunities for the CIO of the future.
Dev Center 20/20
How will we organize development centers for the apps that will power our enterprises?
IT Operations 20/20
How can you achieve the data center of the future?
What the workforce of 2020 can expect from IT, and what IT can expect from the workforce.
Preparing today for tomorrow’s threats.
Data Center 20/20
The innovation and revenue engine of the enterprise.