Companies Crunching Big Data Winning Competitive Advantage
by Tim Baker, CFA. Thomson Reuters
Too much information
A growing number of companies are using the tools and disciplines of Big Data to transform their businesses and markets. Its application in the consumer sector is becoming well known, but it is also playing a crucial role in financial services.
There are many definitions of Big Data but most will agree that it refers to the enormous expansion in the volume, velocity and variety of information available about people and their habits. Much of this data is created as people go about their business online — their digital footprint. When crunched and combined with other kinds of information it can provide marketable insights which were previously unthinkable. A simple example is the use by online supermarkets of analytical tools to recommend to shoppers new products that they might be interested in but were unaware of.
This explosion in data has led to a widening gap in an enterprise’s ability to process and interpret all the information available to it. Perhaps the best indicator of the commercial importance of this field is the growing number of firms providing software and services to help companies benefit from this wealth of information, not to mention the demand for Big Data engineers and scientists.
Big Data also needs to be seen as a means to an end, not an end in itself. Without real purpose, a data project can rapidly become a white elephant. Experts in the company’s business are required to identify the opportunity, pointing to likely data sources to be ingested by the engineers, and to potential relationships to be exposed by the scientists. Big Data is not just collecting information; it includes the need to process it to make it readable — by man or machine — and thereby produce results that can yield new ways of doing business.
So where are the opportunities? Obviously these are generally clustered around where the concentration of data is the greatest — for example in the consumer sector, where social media and online activity are strong drivers. Here the purpose is generally to predict customer behaviour and thus compress delivery times and optimize marketing and price. The data is rapidly changing and requires tools that are capable of generating predictions for very specific needs. One such algorithm is Amazon’s recently announced patent on ‘anticipatory shipping’, a process which uses information about customers to anticipate their needs and moves items they are likely to order to nearby depots, ready for rapid delivery.
Big Data’s crucial role in the financial services industry gets less press coverage but this is a sector that presents unique opportunities and challenges.
For investment companies, creating an information advantage is a way to improve returns. Modern technology has shortened the time it takes to capitalize on such information advantages. For example, 10 years ago a retail analyst covering Home Depot, a home improvement retailer, might have needed to hire tens of people to count cars as they went in and left the car parks of a sample of stores. Now, people have been replaced by satellite imagery, while data from mobile phone towers and locationaware smart phones can provide another level of accuracy. Overlay this geo-mapped data with the geographical footprint of the retailer and you can generate a reliable predictor of sales long before the company makes them public.
Collecting, cleaning and combining this information are not simple tasks. For example, geographic data might come in daily or even hourly but must be mapped to quarterly earnings for a company with a different geographical footprint. Speed is crucial: the faster a company analyzes data, the greater the information arbitrage.
While some departments at financial institutions worry more about return, others are concerned with risk. In a bid to reduce systemic risk, regulators have been clamping down on large banks, requiring them to stress-test their balance sheets against a range of adverse scenarios. Every financial instrument needs to be valued and the interaction between these securities has to be properly mapped and understood. Data feeds are now a key part of these complex models. So important is their use that many firms have appointed Chief Data Officers to supplement Chief Risk Officers.
Data from such sources such as Thomson Reuters helps all kinds of enterprises in their quest for profit and management of risk. What can be done with analysis of data seems almost limitless. What is known as ‘data exhaust’ — the information captured from the activity of internet users — is far from being a waste product. Owners of this information are constantly seeking new ways to profit from it.