Big Data and Analytics – Suggestions to Approach

There has always been an opportunity to create big data. There have also always been opportunities for prediction based analytics in the repeatable processes. For example, a simple thing like driving a car for a mile can generate tons of data, like the oil temperature changes, engine sound profile, traffic encountered, weather conditions faced, break/acceleration usage during the drive, road conditions and so on. Similarly, one can crunch data and keep developing prediction models on what to expect during the drive, like how much time it would take for the one mile drive, how much of gas is expected to be burnt, level of stress expected on the driver during the drive and so on.

In his article, “Big Data is Just a Fad,” Buck Woody concluded, “Big Data…will fade, over time, into the pantheon of other tech buzzwords. But the data it represents won’t – it exists now, and continues to grow. So it’s OK to allow the term for now, learn the concepts it presents, and bake it into what you do today. Big Data will only get bigger. And that’s not hype.” While it is easy to agree with him that tech buzzwords have always had a life of their own, it is just now that means to capture, store, process and analyze the data at every possible opportunity began to unfold with the new inventions around big data and analytics. It is hard to figure out where to start and how to approach using big data.

Here are my three suggestions for organizations wanting to take advantage of the world of big data and analytics –

a) Choose your battles – like in the example of driving a car for a mile, there has always been an opportunity to make data big in every aspect of life and business. Therefore, the most important thing is to pick the right scenarios suitable for the data and analytics.
  
b) Evaluate the opportunity cost vs. benefit – the cost of end-to-end processing of data vs. the expected benefit done the old fashioned way needs a more integrated approach. Often the real enterprise information integration challenges are downplayed during the sleek prototyping and exploration phases. The existing data assets need to integrate with the new unstructured data processing engines. The people, process, tools and technology landscape might need an overhaul.

c) Have a clear strategy – investments in the big data space need a clear strategic outlook. Lack of clear vision and strategy would ultimately lead the organization to a lot of data debt or clean up issues. It is very important to hand hold the organization as they prepare to dive into working with the new technology with a clear vision and focus.