Decisions, Decisions
We spend our lives, both personal and professional, making informed and uninformed decisions all day long. Some decisions have seemingly no consequences while others have long-lasting repercussions.
- Do I want Thai or Japanese for lunch?
- Should we make targeted offers to customers that have been with us for more than 2 years or 5?
- How do fix defective devices more efficiently?
Although sometimes not making a decision is worse than making the wrong one, we all strive to make the best decisions possible. And to make the best decisions, we rely on experience and whatever information is at hand. With experience in the subject matter, we can make decisions quickly. However, when the matter is new or information is scarce, we usually need time to evaluate different possibilities. In business, often a large number of decisions need to be made in a short amount of time. Automation can help speed up the process. However, how can we make informed decisions at scale?
Historical Data is Unavailable
Let’s say you’re launching a brand new product across your stores. Should you stock more in Massachusetts or California? When there is no historical data, decision-making relies on experience, analysis of the situation, and intuition.
- Experience corresponds to a number of facts acquired over time that forms a consistent body of knowledge that can be applied to similar new experiences. This body of knowledge often resides in people’s heads. When formalized, this body of knowledge can be found in policy manuals.
- Analysis of the situation refers to careful study of what information is currently available.
- Intuition is the “gut feeling” that we get when we know that a certain decision or action will achieve a desired outcome more often than not.
All of the above can be combined to gain insights that inform decisions. When verified, the decision-making process can be formalized into procedures or business rules that can be continuously refined as more decision outcome data comes in. Decision modeling is a method to describe the decision-making process and DMN (Decision Model and Notation, an OMG standard) is a common notation for documentation.
Historical Data is Available
Let’s say you’re own a bakery and you noticed that your strawberry muffins sold quite well over the past year but not as well as your blueberry ones. What should you do to increase sales of your strawberry muffins? When historical data is available but the data set is small, like this example, you may be able to derive your decision-making steps quite easily by reviewing your data in a spreadsheet where you can create aggregations and run calculations.
However, let’s say your a manufacturer and global distributor of muffins. When the data set is large, leveraging Business Intelligence (BI) tools will help you analyze your data more efficiently and effectively. Often, these BI tools come with graphical representations that enable you to easily visualize the data. In addition, they come with filtering and other data manipulation capabilities that allow your to “slice and dice” the data. With insights, you can then make decisions. Similar to when historical data is unavailable, you can formalize the decision-making process into procedures and business rules through decision modeling in DMN.
In Part 2, I’ll discuss how predictive analytics or even prescriptive analytics can be injected into the decision-making process.
Learn more about Decision Management and Sparkling Logic’s SMARTS™ Data-Powered Decision Manager