Moving from BI to Big Data
Date: Monday , January 02, 2017
Over the last 3 to 4 years, there is a sudden surge in the trend and shift in focus on evangelizing and monetizing an organizations data. Every role in the IT industry, starting from a newbie in IT to the CEO of the company are talking about deriving more insights using current and historical data.
Why this sudden trend now? On further analysis, this trend is much more in the organizations which have traditional data warehouses (EDW) in place. The reason being, business intelligence teams are utilizing this data more as an enterprise reporting based on the historical trends to derive current business performance. Using this past data, the teams are not able to predict beyond a certain limit for bringing major positive impact to the business. Real time decisions are very tough on the structured data that is coming from the BI Applications.
Based on the recent survey done by the analysts, every day, we create 2.5 quintillion bytes of data so much so that 90 percent of the data in the world today has been created in the last two years alone. This data is more voluminous and wide variety in nature. Unless, we don\'t effectively mix and model this data with our existing applications, it is impossible to predict the patterns for the next best business action. Hence, the need came now for putting more efforts to derive meaningful insights by performing deep dive on volume, variety and velocity of data.
Below is the interest shown in Google on these topics across regions for the period 2004-Nov 2016. The numbers on y-axis represents, search interest relative to the highest point on the chart for that period. A value of 100 is the peak popularity for the term. A value of 50 means that the term is half as popular. Likewise a score of 0 means the term was less than 1 percent as popular as the peak.
Necessity for Aligning to New Trend
The question arises that, when every organization is running towards new technologies, do we need to synch with them? The answer is YES-to meet the competition, to understand more on what is happening and what will happen if the same trend continues and who will get affected and how will it influence my business.
The bygone era of thinking that, the database is nothing but storage of numbers in the form of rows and columns is gone. The new realization is that, Videos, Text, Audio, Maps, Photos, Emails, Chat from various social media sites will also provide lot of valuable insights. To summarize, deep analysis provides any organization with more marketing muscle and in turn, make their customers happier, help their organizations be more efficient, and keep the competition at bay. In Banking, functional areas like Risk, Compliance, Fraud, NPA and Calculating Value at Risk can benefit greatly from these analysis to ensure optimal performance, and to take crucial decisions where timing is very important. The right tools can even recognize specific patterns based on predefined criteria.
Identify Key Stake Holders
Once the buy-in comes from the leadership team, work closely with business teams to find a current business problem where you believe you can have a measurable and meaningful impact. Define your key metrics, gather the appropriate data from various sources, and then iterate through the analytics quickly to find predictive patterns.
Find Right Team
Most of the companies start utilizing their existing IT teams to transform or integrate the data that is coming from the newly identified sources. This will be like, trying to cure a chronic ailment using self-learned home remedy. The key team members for any big data implementations are - Business analyst (Industry experts), Tool expert (Statistically minded and experienced in building models), Data engineer (to transform and process the data).
It is better to leverage existing infrastructure and start with a small use case and a POC rather than, going with a big bang approach. The advantage being, early user adoption and start to lay the groundwork for building a culture of analytics-driven decision-making. It\'s a process of Renew-New (Augmenting the existing BI).
Test, Compare and Repeat the Process
With each data set, validate your assumptions and optimize further. Share the results with business team and improve operations. Create repeatable process and action paths. Finally, if the existing EDW has clean data with proper data governance structure in place and providing single version of truth, it will be little easy to integrate additional unstructured data sources combining with right software and hardware. Providing immediate access to the right data, in the right format, is no longer an aspirational goal - it\'s a basic necessity. Building of an EDW to drive a best-in-class BI strategy is a big step in the right direction. Combining with real-time operational business analytics is a huge hop in delivering game-changing differentiation, innovation and performance.