In today's business environment, greed for data is good. Across the world, businesses have begun to stockpile data. They understand that data is their new asset. As a consequence, data from a variety of sources such as sensors and cameras (industrial, medical, public safety), transactions (ATM, credit cards, shopping), online exhaust (website interactions), social interaction, open data (weather, public transport schedules), historic data (demographic profiles, educational records, insurance claims, immigration archives, etc.), production and enterprise data (suppliers, employees) are being captured. Research firm IDC has predicted that the Big Data market will grow revenue at 31.7% a year until it hits $23.8 billion in 2016. This number excludes revenues from analytics, SaaS applications and products that pivot and leverage data.
So now we have tons of data crawling out of every nook and crevice of business. And it is making its way into sophisticated data warehouses. We know that all this useful data is sitting in our backyard – or in a handy Cloud – waiting to transform business and the way we work. But there is a problem: the data is in various formats, the database has problems of consistency, availability or partition tolerance, and you discover that you are looking up one key to acquire another key to acquire the next leading to accessibility issues when it comes to lightning fast analytics. And 'lightning fast' is what you really need in a business environment like retail which is 24X7, across multiple channels and different geos where customers expect instant responses. What you need is a way to unify the data as well as bring you analytics strategy closer to it.
What is your data analytics strategy? Does it include acquiring Data Scientists to ply through mountains of data to deliver business insights? Or does it include pre-fabricated and re-usable algorithms and frameworks that anyone in business can use to extract insight?
One of the biggest barriers to leveraging data has been the tools and skills required to extract insight. These tools and skills have lowered the ROI on data, made adoption of these technologies difficult and in several instances, despite significant investments, the complexities involved have led to failure. To us, simplifying the data-analytics relationship represents the first step to unifying your Big Data and analytics strategy.
Just as business users could deploy spreadsheets to manage their data – without intervention from the IT department to run the spreadsheets – can your teams deploy analytics engines to manage Big Data? Can the talk of data connectors, clustering, in-memory computing, R code and predictive modeling be minimized? In short, can the CFO and his team monitoring costs or the marketing team addressing new customers with fresh campaigns bring together data from 20 different sources in multiple formats – at rest and in motion – at the click of a button and publish actionable decisions with confidence across the enterprise?
The point here is that Big Data and analytics strategy need to empower your team – from back office to check-out counter – without having to wait for the gurus of analytics in the enterprise.