1.1 Why Data?

One of the first things that an aspiring manager or consultant has to learn about solving a problem is that he or she is not being paid to provide unsubstantiated beliefs about what might or might not be a good solution. Rather, the successful manager or consultant is paid to propose solutions based upon pertinent data and a well-reasoned analysis of that data. The manager’s feelings or guesswork or intuitions can be helpful in exploring a problem situation but they cannot by themselves be the basis for making sound and reliable decisions. Again, it is having a clear idea of how to define the problem and having a plan for collecting the relevant data that constitute the professional approach.

The first step to solving a problem is to define the problem. This is not as obvious or as simple as it sounds; there are numerous case studies showing how businesses have wasted large quantities of money trying to solve the wrong problem. Listed below are the other steps in the general problem solving process. Keep in mind that the process is usually not sequential. You will usually find yourself jumping steps and repeating steps in an attempt to refine your solution.

1. Problem formulation stage
   1. Define the problem
   2. Identify possible causes and their effects
   3. Determine data to be collected
2. Data collection stage
   1. Determine what the variables are and how they will be coded
   2. Construct data collection forms
   3. Construct the database for analysis
3. Solution development stage
   1. Interrogate the data
   2. Determine a root cause for the problem
   3. Develop possible solutions
   4. Use the data to select the best solution
4. Refinement stage
   1. Test the solution with sample data
   2. Modify the solution based on the tests
5. Implementation stage
   1. Present your findings and your plan
   2. Put your solution into practice
   3. Collect data as to the effectiveness of the solution
   4. Modify the solution as needed, based on data

One of the reasons that it is vitally important to define the problem you are studying is because real-world problems are often multifaceted. Their causes may be well hidden, and what you observe - the perceived problem - may mask the real problem’s causes. Part of your job in studying a problem is to think of possible causes for the perceived problem, then determine ways to investigate the situation by collecting data that can sort through these causes. Making this even more difficult is the fact that a single cause can have multiple effects, each of which may generate more effects of its own, some of which may overlap. Identifying this chain of cause and effect is really what understanding the problem is all about.