CNV3-I2-2- Facets-Analytics: Trends and Utility in Real World

Introduction

“Information is the Oil of the 21st century and Analytics is the combustion engine” – Peter Sondergaard, Gartner Research.

As per the Merriam-Webster dictionary, Analytics is “the method of logical analysis”. In simple words, Analytics is the discovery, interpretation, and communication of meaningful patterns in data with the aim of effective decision making. About a few decades ago, communication channels were few and slow hence decision making was also relatively slow. As the digital age set in, the speed of communication increased, and this impacted every aspect of life. Fast decision making became increasingly important and it was also supported by the explosion of data. In today’s competitive world, one cannot expect to succeed in any venture without harnessing the power of data in an effective and timely fashion. This is where Analytics plays a very important role. In today’s world, Business Heads use Analytics to not only analyse what has happened but also predict what will happen in their world and how it will impact their business.

Analytics has many flavours based on Function (Supply Chain Analytics, Retail Analytics, Customer Analytics), Objective (Predictive Analytics, Prescriptive Analytics) or the Techniques used (Big Data Analytics, Text Analytics, Speech Analytics).

Trends in Analytics

As you would see, there are multiple facets to Analytics. A few of them which are currently trending are as follows –

Artificial Intelligence

Artificial intelligence (AI) is the science aimed to make machines execute what is usually done by complex human intelligence. AI and machine learning are revolutionizing the way we interact with systems in our daily life. Day to day examples are the Chatbots or Digital Assistants which are present on several websites. They do away with the need for having real humans interact with website visitors. The interaction with the Chatbot is so smooth that in a few cases, it is virtually impossible to identify if communication is happening with a real person or an artificial one.

Predictive Analytics - Predictive analytics

is the practice of extracting information from existing data sets in order to forecast future probabilities. Predictive analytics indicates what might happen in the future with an acceptable level of reliability, including a few alternative scenarios. It is a great tool for managing risks, identifying opportunities and preparing for the same as it becomes easy to know what might happen in the future and therefore take timely action. Machine Learning techniques like Classification and Clustering are extensively used for Predictive Analytics.

·         Cloud Analytics – Cloud analytics is a type of cloud service model where data analysis and related services are performed on a public or private cloud. The trend of owning servers and hardware is rapidly fading away as it does not make business sense to keep millions of dollars invested in infrastructure. Cloud providers like Amazon, Google and Microsoft have made cloud adoption very easy and quick. A point to note is that these vendors have not stopped at just hardware. They also have Analytics software as a part of their cloud offerings in the SaaS (Software as a Service) model or the Pay-As-You-Go model. Analytics software is typically expensive. However, this model has made it very easy for even small sized organisations to adopt Analytics for their business needs. Microsoft PowerBI, Tableau Public for Business Intelligence reporting, Amazon’s EMR and Google’s BiqQuery for Big Data Analytics are a few examples of how easy it has become to adopt Analytics.

Real World Utility

It is very interesting to know how Analytics is being used in the real world. I am a part of an organisation (www.ellicium.com) which excels in Big Data, Analytics and Artificial Intelligence and we have helped organisations use Analytics very effectively.

We recently helped a leading telecom company in India adopt Big Data Analytics for having near real time analytics of how their cellular service was working at a pan India level. This is being used to predict customer churn. Using the Hadoop platform, we created a solution to ingest data from cell phones and present results to the Business Heads using tools like Google Maps and AngularJS. This allows analysis of the quality of the Cellular Service, right from all India level to a drill-down area of 1 km X 1 km. It also allows visualization of the way triangulation happens between cell site towers when a mobile phone uses the network.

Figure 1 shows the areas where the mobile network is not very strong. Figure 2 shows the cell site towers, which a particular mobile phone is connected to. Do remember, that all this information is in near real-time and allows the telecom company to take corrective action very fast.

Career Options in Analytics

A question which I am always asked is “I want to do a career in Analytics. What options are available to me”. There is no straightforward answer and the options are multi-fold. At a very high level, there are two kinds of experts needed to implement successful Analytics solutions – people who understand data i.e. the Subject Matter Experts (SME) and the people who can help engineer the solution i.e. the technology people or the Data Engineers.

SMEs or Functional experts are those who have a good knowledge of the business domain. Example of business domains are Banking, Insurance, telecom, Healthcare, Manufacturing. It is helpful to have a MBA degree with one of these specializations. However, a point to note is that though a MBA degree will give a good start to an individual, it cannot guarantee a successful career in Analytics. This is because most MBA degrees just give an overview of the domain without going deep into the subject area. As a result, it is essential to get real life experience in the particular domain to become a SME. Also, there are several professionals who become SMEs without an MBA degree by virtue of hard work and spending dedicated time in a particular domain.

Data Engineers are the professionals who have a good command over converting business requirements to software which helps Analytics e.g. Business Intelligence Dashboards, Reports, ETL (Extract, Transform, Load) programs, Data warehouses and other batch processes needed for Analytics. They have a good command over programming and usage of BI tools like PowerBI, Tableau, Qlikview, Pentaho or programming tools like Python, R, SAS, SSPS and Java. The scope of data engineers in India to work with includes organizations in banking, policing, fraud detection, healthcare, telecommunications, e-commerce, energy and risk management.

Yogesh Kulkarni, Chief Technology Officer (CTO), Ellicium Solutions Private Limited, Pune