How big does big data need to be before it is valuable?
The value in big data is within reach of everyone. It could mean wanting to mine a couple of extra fields about the customer or wanting to improve the customer profile using unstructured data about customer interactions using Hadoop. Most articles and hype about big data surround the three Vs; Velocity, Variety and Volume. However, we should never lose sight that big data is relative to your business plan. The real conversation to be had is about the value in being nimble.
The 4th (V)alue: The intersection of big data and high performance analytics
High performance analytics is the next generation of analytical focus as we work our way through the era of big data looking for optimal ways to gain insight in shorter reporting windows. It is all about getting to the relevant data quicker and delivering that information in real time. High performance analytics is equipping David-sized organizations with the tools to level the playing field. Examples include:
- How a bank determines credit risk assessment in seconds instead of hours.
- Where a government agency improves social welfare by analyzing unstructured citizen interaction data.
- An insurance company that uses census data to improve marketing response rates.
- How an online business analyzes social data to understand sentiment, and behavioral data to improve campaign targeting.
Regional healthcare provider and an insurers point of view
I recently listened to an Australian healthcare customer discuss their version of big data and high performance analytics. It went like this.
Through some acquisitions we have increased our data base size by approximately 15 percent. This has resulted in our marketing teams being frustrated with longer than usual time-to-market for gaining customer intelligence and executing campaigns. Further compounding the issue is the competitive pressure coming from recent changes in government legislation, which is driving customers to shop around. This increase in competition means marketing needs to be more nimble. Meaning more campaigns to fewer people with more relevance.
Another example is a local Insurance customer I met with to discuss their version of big data, high performance analytics and real-time analytics.
We have issued our sales force with iPads. The challenge we face is, how do we deliver intelligence to our sales representatives in a manner where we know it is relevant, timely and contextual? We know they are meeting with prospects and customers but how do we analyse customer data, analytical data, transactional data and interaction data to provide a Next Best Offer in seconds?
If we understand how to beat Goliath, do we know what to beat him with? A high performance approach leads us to think about the problem differently and look for a solution that optimises the analytical jobs and the way they were architecturally executed. I expect there is a target value proposition heading my way, now.
Under the hood: high performance analytics is not that scary
We often think of new technology as being like a Ferrari, always thinking it is out of reach or too complex for the average David. The reality is high performance analytics provides various approaches that span the spectrum of your maturity and size, from:
- Moving existing analytical models into operational processes for real-time decisions.
- Optmizing analytical jobs to leverage your existing in-database power.
- Using in-memory analytics to take advantage of cheaper hardware.
- Building an enterprise analytical platform to drive down TCO while always prioritizing business value using a grid based approach.
- Visually exploring big data using high-performance, interactive, in-memory capabilities to understand all your data, discover new patterns and publish reports to the web and mobile devices.
The democratisztion of analytics, especially high performance analytics has allowed every company whether Goliath or David-sized to benefit from big data. Over the next few weeks we will be discussing the impact of the intersection with big data and high performance analytics. In particular providing examples relevant to the world we live in left of the date line. Join the discussion to find out what the innovators are doing and lessons we can learn locally. You can see some more examples here.
Question: What is your big data opportunity? Tell us in the comments below.
Reprinted from Left of the Date Line