Recently, a group of analysts were asked this question: “In your opinion, if you had to pick only one, what is the best analytics tool in the marketplace?” My answer was Humalytica—the human mind. Now, understand that the forgoing is true, except that I developed the name “Humalytica” about two minutes ago. The next portion of the argument is in part the author’s imagination.
“No”, say the moderator, “What is the best analytics workbench that humans have developed?” Responding, I say, “Oh. That’s easy. It would be Debesh Choudhury, Ph.D. After all, his mother and father did develop him.” “No”, says the moderator, “Is it SAS, SPSS, R, Tableau, Informatica, etc.?” “Oh” I reply, “Debesh is rather SASsy, but you R not going to tell him I said that.”
“Okay”, said the moderator, “This is not getting through your thick skull. Debesh is old, I mean really old. What is the best new analytic tool that humans developed, not including Debesh’s mother and father?” “Oh.” I respond, “I see what you asking now. That would be Adam Miller. Debesh’s mom and dad had nothing to do with that, or at least I am 99.97% sure they did not. And Adam is young.” The moderator responds, “Jeff, you are as stubborn as a mule.” I am talking about a tool you use on a computer.” “Oh,” I reply, “I have seen Adam use a computer, but I have never seen him on one.”
Of, course 90% of the foregoing argument is fictitious. However, I stick with my answer. The best analytics workbench in use today is Humalytica, though I admit it has much less usage than SAS. Choosing a tool and calling it the best is backward. We need to define a problem, wisely choose an analytic solution technique (e.g., classification trees), and then choose the best tool for solving our problem using the selected technique. There is no best tool!
Let me qualify the statement, “There is no best tool.” I believe this, but often we have to use the tool that the company has licenses and infrastructure for. So, if my company has SAS and the infrastructure that goes with it, then SAS may be the best tool to implement my analytics solution technique. “But,” you say, “R is free! Why not use it instead?” Well, the answer to that is twofold. First we have the infrastructure for SAS. All the company’s data is in SAS databases and the company has invested time, effort and the almighty dollar in building and maintaining that infrastructure. Second, I do use R to test my SAS results.
However, the argument over SAS or R, Tableau or BOARD, QRS or XYZ, etc. is the wrong dispute. A more appropriate squabble is did Humalytica select the most appropriate analytics solution technique? The tool-centric argument is pointless, except to the extent that budgets usually constrain a company in tool selection and number of tools to invest in. So, the moderator had a legitimate question, assuming he was asking, “What is the best tool we should invest in?” In that case, it depends on budget, not just for the software licenses, but the infrastructure as well. Then, which is the best, given we have the budget for one analytic tool and its associated infrastructure? That is easy. It is a combination of Humalytica and the most flexible tool we can afford!
Jeffrey Strickland, Ph.D.
Jeffrey Strickland, Ph.D., is the Author of Predictive Analytics Using R and a Senior Analytics Scientist with Clarity Solution Group. He has performed predictive modeling, simulation and analysis for the Department of Defense, NASA, the Missile Defense Agency, and the Financial and Insurance Industries for over 20 years. Jeff is a Certified Modeling and Simulation professional (CMSP) and an Associate Systems Engineering Professional (ASEP). He has published nearly 200 blogs on LinkedIn, is also a frequently invited guest speaker and the author of 20 books including:
- Operations Research using Open-Source Tools
- Discrete Event simulation using ExtendSim
- Crime Analysis and Mapping
- Missile Flight Simulation
- Mathematical Modeling of Warfare and Combat Phenomenon
- Predictive Modeling and Analytics
- Using Math to Defeat the Enemy
- Verification and Validation for Modeling and Simulation
- Simulation Conceptual Modeling
- System Engineering Process and Practices