Monetising Data

Because of the emergence of the IoT technology the humongous volume of data will be generated and stored in various databases at multiple geographic locations.

Monetising the data is a key strategy for businesses today !

Most of the big companies already have implemented strategies to monetise their data. Some of them offer Data as a Service (DaaS) which is the new coined term for APIs (Application Programming Interface). Traditionally companies provide access to their data/data product using APIs through which other applications interact with their product or use their data.

Typical example is Google Maps where Google monetises the Geographic data they have collected over years. Even though Google Maps is their product, anybody can use their product in their business products/applications through Google Maps APIs. Google charges you for accessing their data through their APIs. (Facebook actually buys a company if they are getting more hits for a particular API !)

How about the private organisational data ? How do we mine intelligence out of the data to get some business insights? How can be use all the past data to forecast sales better? How can we use CRM data to understand customer problems and deliver more value? How do we improvise our product and business processes?

Business questions are driving the data analytics now and data scientists need not only data analysis expertise but also business knowledge to translate business problems into data analytics problems.

For example if the business wants to know profile of customers who can be offered some premium products, then the data scientist/analyst needs to understand the business requirements and they can pick the “RIGHT” variables and data sources for doing this “classification analysis”. Gaussian Mixture Models / Hidden Markov Models can be used for classification analysis using right features and variables.

Data scientist/analyst should possess strong statistical modelling skills and should understand the basic assumptions involved in implementing any data analysis tools.

If the data analyst wants to perform correlation analysis to understand the influence of variables, they have to know that correlation works with data which has GAUSSIAN distribution. Underlying assumption in the regression analysis is that data is linear ! (In practical situations data is never linear !!)

To summarise monetising data is the next growth area for the industries and data scientist/analyst play a vital role in that business strategy.