What is Data Science?

Data science is essentially the study of how to collect, process and interpret large sets of data. However, the purpose of this article is not to focus on what data science is but more so on what data science does or can do in your organization today.

Customers are more active than ever and this means that businesses have the opportunity to capture more data than ever about what’s going on with the customer. By leveraging data, companies are now able to do things they’ve been wanting to do for years in the hopes of gaining an edge on the competition. Today’s companies are learning to use data to be more efficient in their operations, more reactive to changes in the market, to quickly identify and resolve problems even before they happen. Through all of these and more, understanding the value of data and how to properly analyse it will produce benefits for today and many years to come.

Firstly I would like to address the elephant in the room, does full mastery of Data Science require you to be a learned physicist or statistician or programmer? The answer to that question is not anymore. Technology has brought us to a place where there a host of tools available such as RapidMiner, Microsoft Azure and Google Cloud Auto ML that are helping make it easier and easier to take the leap into the world of data science.

So how do you know if data science can benefit you? Let's start by answering this question. Are you currently working in the Energy, Finance, Gaming and Entertainment, Hospitalities, Government, Health care, Insurance, Technology, Manufacturing, Pharmaceutical, Retail, Telecom, Travel and Transportation or even the Utilities Industry? If you've answered yes to any of these, data science is definitely for you. The power of data science is that it doesn’t matter what industry you are in the same benefits can apply to you because data and technology still apply to every industry or size of business.

In October I attended the 8th staging of the RapidMiner Wisdom Conference where I was able to see some of the great advances being made by companies in various industries from all over the world all using a form of data science called machine learning. Here are three examples of the many innovations that were on display:

1) Lufthansa Airline

Lufthansa Airline set out to accurately predict the arrival time of aircraft once they were air-bound. The ability to predict accurate arrival times are crucial for a vast number of airline operations. Such as Requests for priority landing, aircraft changes, turn around management, crew change over, flight delays, rebooking of passengers, and most importantly overall customer travel times which directly affects customer satisfaction. All of which correlates to specific revenue and expenses for the airline.

2) EzCater

EzCater used data science to uncover the issue behind ineffective marketing campaigns focused on the Customer Lifetime Value. What their data models were able to predict is that the estimated year 1 Revenue of a customer was more significant than the overall Lifetime Value in determining the types of discounts and promotions to offer to customers. Using a mixture of historical data and predictive estimates they were able to predict the overall value of a customer at the end of 1 year.

3) Lindon Ventures

Lindon Ventures were asked to solve an ongoing problem on behalf of a client. Their client owns over 100 call centres each of which receives on average between 500-1000 calls per day. At this volume, the client wanted a way to identify calls with repeated problems so as to ensure the correct actions could be taken: training for staff, customer call-back or problem escalation. The solution presented was text mining using the transcripts of incoming calls. Through a mixture of filtering for repeated words, exclusion of stop words, and assigning weight to specific call attributes. Lindon Ventures were able to develop a machine learning approach that predicted and monitor the trend of "bad" incoming calls which enabled their client to anticipate and reduce escalated customer service complaints.

Conclusion

It is important that when starting new initiatives such as these that a highly qualified team be used to ensure effectiveness at each stage of the project. Resources will be needed to ensure business value as well as support any infrastructural needs but there are two specialised roles which can add tremendous value. A Business Intelligence Expert is required to ensure that your solution is designed to meet your strategic, operational and performance requirements and that robust mechanisms are created to ensure data is properly extracted from various sources to the solution. A Data Scientist can be a valuable asset to provide guidance into designing and implementing data analytics as well as identifying the data elements which can provide the most value. It is not to say that success cannot be achieved without data science, however, the question is why tackle the future with the strategies of the past?