This week we have another guest post by Abhishek Mittal, our Jigsaw-Bocconi EPBA (Executive Programme in Business Analytics) student. In this post, he talks about how analytics should be implemented in the practical sphere.
So you have heard by now that data is the heart of many businesses. That’s right, business world has changed and data analytics is a widely accepted phenomenon. Oh really is it? In that case I will also implement analytics in my firm. Yep that sounds like a good idea but unfortunately it is not that simple and takes a different evolutionary path altogether as compared to any IT or other business process framework implementation. This article talks about the strategy and approach to implement Analytics in a company.
The virtuous cycle of data science is an iterative learning process that builds on results over time. No doubt, success in using data will transform an organization from reactive to proactive. However, there is lot of analysis to be done to decide upon how to go about implementing analytics from scratch.
To start with, one needs to understand the Information Continuum that exists within the company, Data warehouse capabilities, ETL processes, data entities, business processes and the existing decision strategies.
Remember you do want to leverage the existing data warehouse to establish the analytics practice. You have already made the huge investment in DW and DW professionals in the company already understand the data entities and ETL processes in depth.
With the said approach not only you will incur reduced costs by not reinventing the wheel but also the analytics framework you are designing is well aligned with the existing business processes. Next you need to map the different data flows to the business flows to get the holistic understanding of the business. Next you create an Analytics DataMart in the DW with the help of DW professionals and by leveraging the existing ETL processes. Of course you can introduce new ETL processes depending upon the needs given that you already have the tools for it. Analytics DataMart is the place where all the analytics inputs and the outputs are stored.
As a next step analytics framework should be integrated with the existing business operations systems in a loosely coupled manner. For ex – a separate event is generated for a specific use case and then propagated to the analytics framework, which runs analytic models (established by analytics professionals for that use case) on the input data. The output from the Analytics framework triggers the automated decision based on the decision strategy and rules. Finally, the meta data (operations metadata, analytics metadata and decision metadata) for the entire use case(s) should be stored for audit and control purposes. This is important to measure the value added by analytics and also for continuous improvement as the data or model may change with changing market dynamics. See figure below:
To summarize, following steps should be followed diligently to implement analytics practice successfully:
Implement Analytics the right way and you are on your way to untold success.
Abhishek Mittal has over 15 years of experience in designing and developing enterprise applications using technologies such as Big Data, Analytics, Solr & Elastic Search, Cloud Azure and AWS, .Net, IOT and Mobile. Bitten by the analytics bug, he has enrolled in the Jigsaw-Bocconi EPBA (Executive Programme in Business Analytics) course and is looking to use analytics to solve real-world problems.
Other articles by author