While I am not a calculus guru, I do work for a consulting firm that is more than happy to hire math nerds and capitalize on the needs of businesses. There is a saying in consulting that “if you’re not part of the solution, there is good money to be made in prolonging the problem.” That is only partially true. In fact, consulting firms recognize opportunities and make themselves indispensable. Where does math factor into this? The answer is through data mining.
Data mining can sift through endless piles of data and identify trends. This predictive function is vital to many organizations. Furthermore, a good data mining system can discover these trends through automation. As a result, the company needs only to have personnel on board who can analyze the results. The math geniuses who designed the program are viewed as the conquering heroes. Of course, if the company does not employ anybody who can analyze the data, my firm would be happy charge the company for these additional services.
Companies that are not strong in terms of technology can easily fall into not knowing what they do not know. This is a weakness that perpetuates itself until poor results make themselves blatantly obvious. If leadership believes that maintaining an Excel spreadsheet is going to provide a reliable system, then they are mistaken. Instead, management must look at how they approach data mining. First, objectives must be clearly defined. The process of developing data mining infrastructure goes hand-in-hand with strategy. As a strategy consultant, this interaction is vital. Great strategy means little if there is not a system to evaluate the data it produces. On the other hand, great modeling programs mean little if the input or output is unknown or undecipherable. Once the company defines how it plans on using the data, the correct information can be categorized.
Should a client present itself with a data mining need, there are steps my firm would take to provide results. The strategy consultants would determine how the client is positioned in the marketplace and work with leadership to discover the way forward. This involves marketing statistics and data that is either available but cannot be trended, or has yet to be discovered. Preparing the data is often the most time consuming aspect of this process.
Next, we would determine what model to use. This depends not only on what industry the client serves, what its goals are, and its current capabilities; but also how much they are willing to pay for a sophisticated product. Once this is recognized, the technology team will utilize math and logic to develop a vehicle that will enable the data to be gleaned into relevant information. Market analysts can assist this process by making sure that the scope of the data mining meets the needs of the company (i.e. data quality).
Finally, if needed, consultants would interpret the data. This would enable the company to clearly see where they are weakest and identify trends. For example, data mining is great tool to use for SWOT analyses. The game is not over at this point, though. If predictive trends aren’t realized, then the system is not successful and has little value to management. Analysis must be done to scrutinize how the organization can either utilize the data to cut costs or take advantage of new opportunities. Also, there is the question of who officially owns and maintains the system.
A final consideration is that data mining is dynamic. Once the desired goals are realized, there is a necessity to continue adapting. While data mining statistical trends are largely historical, they still require a consistent flow of new information.
No comments:
Post a Comment