Of course your decisions should be data driven! Everybody knows that! Right…? And you’ve got BIG data to drive those decisions! And it’s gonna’ make you RICH! RICH I tell you!!! Right…? Everybody says so….
We are in an interesting time, where we have a lot of impressive tools and techniques for managing data-driven decisions, which we should be able to apply to a wide range of problems, but very little clear understanding of exactly how to do that. We (yes, all of us, unless you are an off-the-grid survivalist) are generating unimaginable quantities of data, on everything from where we go and what we like to what our brain waves look like when we’re dreaming. And we intuitively think we should be able to use that data to make better decisions about everything from marketing to health care. And there are a lot of vendors ready to sell you exotic software based on impressive-sounding algorithms (admit it, you get excited when somebody whispers “Multivariate Adaptive Regression Splines” in your ear) to help you do it. But do you really have any clue how to make a decision using your data and any of those tools or algorithms?
I work for FICO (one of those vendors), and yes, we would love to sell you our tools and algorithms. And I do believe we’ve done a lot to make those products more usable by more people, and more applicable to a wide range of problems. But what I’m wondering is how folks who have never worked with decision management tools such as “predictive analytics,” “business rules management systems,” or “optimization models” will figure out how to recognize when and how those things would be useful.
I'm planning on doing a series of these blog posts that I hope will orient the uninitiated into the exciting and profitable world of decision management. (If you are more altruistically-motivated, you can substitute "beneficial" or "empowering" for "profitable" in the preceding sentence--it's all good as far as I'm concerned.) To get started I'd like to pose some questions:
- Does the phrase "decision management" mean anything to you (if so, please elaborate)?
- If I suggested that you "model your decision" would you dive right in (if so, using what tools/techniques/methodology?), or move on to a more pleasant topic such as sports or the weather?
- Do you consider yourself a nerd? If so, would you classify yourself as a:
- Data Nerd
- Software Nerd
- IT Nerd
- Business Nerd
Your answers will help me focus on topics that might actually mean something to you. The last question is significant because any attempt to make data driven decisions invariably involves bringing some combination of nerd types into contact with each other, which invariably causes communication breakdowns, wasted time, and hurt feelings. These nerd types all speak different languages and have very different perspectives on the decision problem domain. And they CARE (which is ultimately what makes nerds of any variety so lovable)! If I know what flavor of nerd you self-identify with, I can try to direct you to content that is produced in your own language and according to your customs, and can help you fit in to this new and expanding universe. Stay tuned!