What’s your take on this??
Well as a personal opinion, i suppose it is one of the best pairs(‘made for each other‘), the reason being “Business being dynamic is always subject to change and variation” and “Analytics on the other hand becomes handy when there is variation so that a pattern can be depicted!!“. Both are mutually dependent on each other.
- To answer this in a very simple and layman’s term, “Analytics help us to get an Insight and Edge over the Customers by the studying the pattern and variations in the data sets and coming up with some Game Changer options”.
- Whenever a Client starts a business, it continues for some years with development enhancement and maintenance. But as time progresses, the man hour effort gets reduced, and at the same time the functionalities reaches a threshold point where it saturates!!
- But to pace-up with the dynamic trend of Business, some value adds needs to initiated to attract more Customers and create a win-win situation.
- The Best source of Information to study the characteristic and nature of Customer is to play around with the Log/Data dump which has a lot of valuable information and come up with a pattern and analytic solution and have the campaign/offers/goodies etc targeted for these group of customers. Its more like understanding the Lifestyle, pattern of Transaction and Profile of the User, so as to have the Targeted Cross Sell Products aligned with it.
- On a realistic note, it does not make sense to bombard Offers via email to all customers (for eg: College Scholarship offer wont hold good for a Bachelor but it might look promising for a person who has a son/daughter going to college).
- On a lighter note , for example, it really does not make sense to have a “Christmas Offer” during Summer. J
As people say it “Right Information…to the Right Customer…..at the Right Time”
How to get the Right Information ??
- It’s a very false notion to consider that we can start with the Analytics on the very 1st Day of the Business or at the Nascent State.
- “Business Analytics” is nothing but the conclusions drawn based on some hypothesis, assumptions, sample data, pattern, time or other attributes that are likely to affect the distribution.
- Analytics is actually the face of the Statistical and Data Analysis. The foundation if built on the Statistical concepts and modeling.
To coin some of the terms and refreshing the old school and college memories here we go :
Mean/Median/Mode/StandardDeviation/HM/GM/Kurtosis/Covariance/ Central Limit Theorem/ Chebeshev’s Theorem/Simpson’s Paradox/Predictive Analysis/Forecasting / Optimization/p-value and many many more…
So, as the underlying concepts of Analytics is purely based on mathematical formulae and theorems, it is very important to get a proper sample/bucket of data to draw a pattern or conclusion out of it.
Methodology to deal with the Raw Data ?
There are different Methodologies too that are followed for dealing with the Raw Data :
- SEMMA SAS (which has its own Data mining Process)
- CRISP – DM (Mostly used one)
Roughly here are the few steps taht are followed as apart of the Methodology :
Each of the above steps has several other sub steps involved, that help us to get close to determining the pattern or distribution.
Business Intelligence vs Business Analytics
Sometimes there is a confusion that pops up wrt BI vs BA. So here is a quick snapshot of the differences.
- Not Really. No doubt the IBM BigSheet (the metaphor Sheet is used to help the customer correlate to big chunks of Data) GBs and TBs of data for Analysis and Filtering purpose, but it has been mostly targeted for the Social Networking and Some other varieties of Data. The Analytics used is very good but it may not fit into all types of Business Domain Implementations.
- We cannot have the Fraudulent and Campaign Marketing analytics done with BigSheets. It is for a sector of users, all usecases cannot be mapped to it.
- There are specialized tools like “SAS” and few more are available for facilitating BA implementations
- So, in order to come up with a Business Analytic System for a Specific Domain, the Requirements and the data Points needs to be correlated so that they can be clubbed for a promising result.
CEP vs Analytics ?
As both of these features deal with the prediction and analysis done on Data sometimes it becomes quite confusing!!..Here you go!!
- CEP refers to Complex Event Processing in real-time for predictive analysis.
- Its like based on certain behavior a trigger should be fired.
- Tool Used eg : TIBCO Business Events
- Analytics refers to the continuous iterative exploration and investigation of past business performance to gain insight and drive business planning.
- It actually helps the business to come up with the pattern and distribution of data based on the sample
- Tools Used eg : TIBCO Spotfire+ , SAS
Business Benefits using Analytics:
- Critical product analysis
- Improved customer service
- Up-selling opportunities
- Simplified inventory management
- Competitive pricing insights
Questions Floating in the Air ??
- What are the gaps in the BPMS and the BI Tools.
- How deep we want t dive into the Business Analytics
- How and where can analytics be leveraged to furnish better results.
- What are the sample usecases where Analytics can pose as a Game Changer
Please do share if you have any inputs on the above scenarios!!
Useful Links :
Happy Learning!! 🙂