Advanced analytics consultant. Analyst. Data miner. Data scientist. Predictive modeler. Statistician.
Call an analytics professional what you will, but all these titles -- even the new and trendy data-scientist label -- essentially describe the same type of individual. And what makes that individual a really great analytics professional comes down to five core traits: commitment, creativity, business savvy, presentation skills, and intuition.
Without each of these characteristics, an analytics professional may be good but not one of the best, according to Bill Franks,
chief analytics officer at Teradata and a faculty member of the International Institute for Analytics (IIA), a research firm focused on how to achieve analytics excellence. Franks shared his views during a recent members-only IIA Webinar, "What Makes a Great Analytics Professional?"
Commitment: The no-brainer among the traits, commitment is easy to see and quantify when talking to job candidates, Franks said. All you have to do is listen carefully to find out if a candidate has gone the extra mile. Someone who is committed might say, "Well, I thought A and B would solve the problem, but just to be sure, I also looked at C and D."
Creativity: "This is one of the ones that's often surprising to people who aren't familiar with analytics talent… but it's very important," he said. "Every business analytics endeavor is different. You'll have a different problem to solve for which you'll have different data, and you'll have to pull it together in a different way -- and it takes creativity to figure that out. In addition, creativity comes into play in dealing with the inevitable data problems. "What do you do when you see outliers or data with errors in it? How do you deal with that with the confidence that the end result won't be compromised?"
Creative sorts are rare among analytics candidates, Franks said. "It's one of the big limiters, with maybe 10 percent to 15 percent passing my creativity mark." To determine creativity, he asks job candidates what they'd do at that "oh no" moment when the data clearly is problematic. Somebody who isn't creative will merely detail technical steps. Somebody who possesses this trait will tell a story around the technical details.
Business savvy: This is the "softer side of the analytics professional" and the second limiter. Franks said business savvy, which is an inherent trait, does not equal experience. It's an understanding of the business combined with the ability to focus on what's important to it. This also involves cultural awareness, especially for globally competitive companies.
To assess business savvy, he asks job candidates why they've made a particular decision, and then he looks for an explanation of business implications and practicality in the answer. "There's a red flag if the answer is just technical."
Presentation skills: "This is huge, but it's amazing to me how many people miss the boat on this," Franks said. "You've got to be able to present results in a way that a) you get the key parts across to the business person, b) they believe and trust you, so that c) they'll take the right action."
To determine presentation skills, he suggested giving analytics professionals a test drive during the interview process. Watch them in action, and be sure to toss in off-the-wall, random questions during the presentation. This is important for gauging how they'll react when, say, an executive asks something ridiculous during a formal presentation. "You don't want to see them wince and look like the person is crazy before answering."
Intuition: This is the hardest trait to define, Franks said. But you know people have intuition if they seem to have a knack for making the right decision, on the first shot, when presented with different options.
Analytics professionals should be evaluated on these five traits, Franks said, and you should always keep in mind that "there's a ton of art along with the science."
Do analysts have to be as much data artist as data scientist? Share your opinion on the message board below.
Many correlations are similar to analysts having to be as much data artist as data scientist. Because data scientists have to be able to deal with mass sets of data this is a skill one should be educated with and have an experienced level understanding of the specific tools. Being able to transfer and deliver to the organisations culture is where that urge for being part data artist can be significant. "Test and Learn" observations is a repetitive process; this process is common formulations within the sciences. The marketing and cross functional departments have been working with big data but still its recently. These areas focus on raising questions that concern the customer. So they are more creative questions based on business outcomes - What are our customers talking about? Is the response positive or negative? When you reference the educational backgrounds of the technical and marketing departments, from the CTO to the CMO and from the Engineer to the Marketing Specialist the backgrounds are from having degrees in marketing to having degrees in computer science.
To a certain extent, then, Cordell, I propose that one of the major jobs for the analytic professional is anticipation -- anticipating what the major pertinent business issues and ramifications of data might be, even if s/he is in actuality unacquainted with them.
All the more reason for clear lines of communication throughout the process, so that nothing is a surprise, and everybody knows what they're supposed to be meeting about.
Indeed, Beth. People like to promote the stereotype of the antisocial IT worker, but in these days, where IT is so necessarily integrated into most important business functions, the successful IT worker is not only anti-antisocial, but knows how to speak the language of the marketing department, the accounts department, and more.
Creative problem-solving is indeed important. This reminds me of a time in middle school when we were to be giving a presentation about metric weights. The teacher made clear that we were to have all of our materials ready in advance; her poor communication indicated to me that she meant "before our individual presentations" -- not before class had started that day.
Consequently, when it was time for my presentation to start, I was flatly told that I was not allowed to gather my materials.
Thinking quickly, I made use of random objects I found lying around, demonstrating the same scientific principles with them.
It couldn't have worked out better than had I actually been "prepared." I got an A, and the teacher was ultra-impressed with my ability to think on my feet.
Nice example Beth. Effective communication is really an essential tool to have if you have to go up the ladder. However, the communication power may be short-lived if it is present without the essential quality of professional skill.
So Shawn, I kind of infer this in another board comment on your ClearStory blog, but I'm wondering what you think about giving business professionals the ability to run analysis on big data without necessarily possessing basic analytics knowledge. Could that be a recipe for disaster should said individual drive decisions based on data he or she isn't really knowledgeable about or capable of assessing in terms of results?
Great point, Beth! As I mentioned earlier, these are traits likely to be found in any business professional and, I would think, especially in entrepreneurs. Since a growing number of startups, not to mention businesses of any kind, will likely be data driven, clearly there is some synchronicity here.