Discover PerformanceHP Software's community for IT leaders // July 2013
Turn unlimited curiosity loose on your big data
A philosophical shift in analytics means you can ask your data anything—and expect an answer.
Analytics is speeding up. It’s now possible to query your data iteratively—at the speed of conversation. But how good a conversation is it? Speed lets you refine your business insights, but you may still be limited by the content of your data, forced to tailor your questions to the structured columns of a database.
Greater breadth of integration is combining with faster data-crunching to let you talk with your data about more things. Combining structured and unstructured information in an open data platform lets you ask nearly anything, such as:
- What’s the lifetime value of customer X?
- How likely is it that a customer will churn?
- When will my equipment fail, and what can I do now to prevent that from happening?
- How can I monitor patients’ daily health to prevent complications from disease or medications?
Five years ago, this kind of curiosity-driven analytics was impossible. Now that the capability exists, it is driving a critical philosophical shift that puts data and analytics at the forefront of business strategy.
Embracing an open data platform
Answering new questions often means making two or more data systems interoperate. The data analytics market is evolving to meet that need, with companies such as HP creating cooperative products that, as parts of a larger data analytics platform, let you capture, store, manage, analyze, and optimize data.
Each of those capabilities is useful on its own, but it’s only when they cooperatively process the same data in up- and downstream processes that businesses can expect fast, reliable answers to whatever they need to know.
If the answers to your most pressing business questions aren’t immediately available to you, the new data philosophy empowers you to take action by adding capable technologies to your data platform. Start by figuring out what your current capability is, and compare that to the potential still locked in your disparate data systems.
Here’s how we recommend doing it: Take a measurable business problem—churn, customer satisfaction, profitability of product lines. Feed all data that you’ve “tagged” or deemed valuable into Vertica for analytics to disprove/confirm theories, and iterate on that data. Much of the exploration to determine if data is valuable is done in Hadoop.
Best of curiosity-driven insights
Asking your data what you need to know, rather than what it can easily tell you, offers immediate and impressive rewards that can significantly affect customer service, infrastructure costs, marketing effectiveness, and overall profit. For example:
- Capacity planning: It’s critical in many industries to build out just the right amount of infrastructure. Telecommunications companies are combining predictive models with location and operational deployment data to know where capacity is under-provisioned—before service to subscribers suffers.
- Targeted advertising: When companies develop richer customer profiles, they can do A/B testing to select the optimal offer to a potential customer, for a better ad ROI. Mobile carriers often do this, using device location data and short- and long-term interests and purchase history.
- Cyber security: Organizations can use predictive models to learn when they might draw greater notice from adversaries. Media companies, for example, can analyze historical attack activity, and then add social media data to help recognize when they’re most likely to be targeted—and how best to take action.
- Fleet optimization: Whether it’s commercial jets or delivery vans, many companies need to take corrective action on service vehicles to minimize preventable delays. Airlines, for example, monitor social media for trending cues (or unstructured data like weather data to determine likelihood of delays), then correlate those patterns with equipment sensor data. This can reveal when equipment is to blame for poor customer service.
Unlocking performance with data
Organizations no longer need to think of data as rigid, finite, or slow, letting them put analytics at the core of everything they do. From keeping networks operating at peak efficiency to addressing customer service issues, from buying advertising to preventing fraud, data is the one asset that enterprises can rely on to improve performance on KPIs.
IT is experiencing a fundamental shift from the “T” to the “I,” with information management becoming the new source of competitive advantage.
Once you know where your analytics tools are falling behind the evolving industry—and possibly your competitors’ capability—you can start to figure out how to get faster answers from a greater breadth of data. Consider the holistic approach you need to make the most of all of your data, but also identify discrete projects that can let you test new tools or methods on a manageable scale.
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.