Advanced Analytics: Seven Steps Toward Adoption
Organizations want to move beyond "rear-view-mirror" reporting, but how can they act preemptively instead of reactively? This report explores the advantage of advanced analytics and helps beginners put methods into practice.
Decisions made on the basis of intuition, visceral fortitude (guts), and years of experience, while valuable, have been proven, on their own, to be less effective than scientific methods. Given the overwhelming complexity and scope of today's dynamic, computerized world, the old methods of decision making are no longer viable.
Wouldn't it be great if there were a methodology that could guide us in our decisions, incorporating the crucial variables that impact each path forward? And what if that methodology would allow us to model and simulate our environment and the impact of our choices before we implement changes that might not work? Analytics is a discipline that enables us to do just that.
More Software Insights
- The Untapped Potential of Mobile Apps for Commercial Customers
- The Critical Importance of High Performance Data Integration for Big Data Analytics
White PapersMore >>
Analytics have been the focus of a great deal of marketing and media hype in recent months, but beginners should keep in mind that it's a practice that requires more than products. This report describes the basics of analytics and outlines a seven-step process that will guide beginners toward success.
What Is Analytics?
In 1911, Fredrick Winslow Taylor published "The Principles of Scientific Management," which spawned the academic field of Operations Research / Management Science (OR/MS). This is an awkward term, to be sure, so today's practitioners increasingly refer to the field using the simpler term "analytics."
Analytics is ideally suited to addressing problems involving complexity and uncertainty. Many people work in fields in which the methods and models they use each day were developed in, or derived from, analytics. Strategic planning, economics and marketing, for example, have analytics at their core. Yet there appears to be a lot of confusion about the role of analytics and its applicability to new areas of the organization -- perhaps areas where conventional BI and reporting have held sway.
A Google search of the word "analytics" (circa February 2010) yields roughly 84 million references falling into three main groups:
- Web-, text- and data-analytics
- Marketing analytics and customer relationship management (CRM) analytics
- Enterprise operations-, process- and performance-analytics .
So, what is analytics? It's the application of mathematics to organizational operations. Statistics, linear programming, queuing theory, network analysis, multi-criteria decision making, simulation, and decision analysis are techniques that are usually at the core of an introductory curriculum in analytics. This is not an exhaustive list, but these techniques are basics upon which more-advanced analyses build. Let's take a closer look at all seven techniques.
Statistics provides methods for describing data, and gaining information from data, using a variety of methods. For example, statistics can tell us, within a range of certainty, the probability that a population with a mean disposable annual income between $50,000 and $75,000 will purchase a specific product within the next business period. Statistics can also then show us the correlation between marketing campaigns and actual sales.
Linear Programming methods are used in optimization, where it is important to maximize or minimize an objective of the organization. Each objective is formulated as an objective function, like a mathematical function, to be solved within a certain set of constraints. Extending the example above, we can use linear programming to maximize the profit of our product, subject to the constraints of manufacturing capacity, the fixed costs of production, and any other conditions that would affect the objective.
Queuing Theory, or "waiting-line" analysis, evaluates factors involving queue arrival, selection, and departure. For example, considering the queue of potential customers for our product, queuing theory can help us determine waiting line bottlenecks or excess capacity in order to manage our points of purchase. This will ensure that the greatest number of customers have access to our product, thereby avoiding customer dissatisfaction and attrition.