Business intelligence, or BI, has become a critical aspect of business management. The ability to pour through data and gain insights can mean the difference between a good business and a great one. That has become only more critical with the advent of so-called big data.
Companies have begun to look to massive amounts of data from internal sources, social networks, data brokers, governments, GIS systems, and other sources to better improve strategy, capital planning, operations, logistics, merchandising, marketing and sales. But there is a major problem: a shortage of data scientists who specialize in wrestling nuggets of intelligence from the mass of names, figures, locations, orders, actions, and all the other types of data that might be available. As Forbes.com contributor John Furrier writes:
Evolving quickly alongside today’s business demands, the data scientist profession is faced with the challenges of interdisciplinary rarity. And the job pool is struggling to keep up. Just as we’re beginning to put Big Data in its place, its human counterpart seeks definition within the modern enterprise. Can the data scientist keep up with today’s growing data demands? The specialized and expensive skills required for this coveted profession are hard to find, making the data scientist a difficult role to scale.
Furrier calls for software automation that can help because “throwing more data scientists at the problem of data management won’t solve” the data scientist’s workload problem. And that is true. There is even software that can help in untangling the nests of data into something useful. However, it is only part of a solution. Automation is fine, but it can lead to some glaring mistakes when people put too much trust in what applications tell them to do. Look at modern spreadsheets. They have data management and comparison, statistics, graphing, and other sophisticated functions to help people make sense of information. And yet, managers who don’t understand the subtleties of analysis and interpretation can make horrendous errors and send their companies down blind alleys, costing their companies money, time and opportunities.
The answer is to educate managers in the need for and use of data analytics. People going through MBA degree programs need a grounding in data. That includes training in data quality, data sources, integration of different types of data, manipulation and interpretation. To assume that all knowledge of the subject should be left to data scientists is like saying that managers don’t need to understand anything at all about strategy, operations, marketing, finance, or other fundamental aspects of business so long as someone in the company grasps the topics. Can you imagine a manager who wasn’t able to type a memo or email, depending on clerical staff to do such work?
Having more managers become smart about data will help relieve data scientists from becoming report machines and free them up to do the modeling and more intricate analyses they are capable of doing. Managers will see faster response times for most data questions because they won’t need to wait for other people to answer. As a result, the company will become more nimble and competitive, being better able to respond to market changes.