This is the first in what I hope is an extensive series of profiles on people who work in analytics. I want to share the experiences and opinions of other professionals so that my readers get a more complete picture. Hopefully my guest’s views will mirror mine, but I will not censor their views when they disagree. I believe it will help my readers understand why there is no broad consensus on many analytical issues.
Tri Ngo has been working in analytics for five years for after previously working in as a project management in the wind power industry and supply chain in the US Air Force.
After working in project management, he considered what the next step in his career was. With the downturn in the economy, the wind power industry was in a contraction. He looked into an MBA, but decided to study predictive analytics at Northwestern instead.
Northwestern is one of the finest universities in the world. The fact that he went there played a big role in Tri’s success. “The knowledge I gained from the analytics program at Northwestern definitely helped me to be marketable to analytics jobs; it has been a part of my career progression.”
Half way through the program, Humana offered a data analyst position which set him up to become a senior analytics and consultant roles at Synchrony Financial-formerly GE Capital Retail Finance and BMO Financial Group. “Definitely, my analytics career started at Humana. I learned so much about database systems, SQL, SAS, Tableau and other BI applications as a data analyst. I was having so much fun applying some of the regression techniques I learned to real business datasets.”
“Going from healthcare to financial services was not a problem. Analytics skill is just numbers and interpretation. At Humana, I analyzed patients’ data to draw insights such as what and why patients stayed at hospital and for how long, what diseases were most prevalent, and drugs utilization analysis. At Synchrony Financial, I analyzed customers spending behaviors, industry trends, customers’ footprints. Beside programming skills, one of the core skills of analytics is to understand the numbers in order to draw findings and insights. Once you understand that and you add in the business context, it is not difficult.”
But there is still much to learn after school and on the jobs.
“School does not have enough time to teach you actual real world programming. For core modeling classes, every week of the quarter, we learned new statistical model, we spent so much time on reading assignments, learning how to write SAS statistical models, and how to interpret outputs.”
“The program provided me the knowledge of what model applied to what case. Actually building those models there are many things I need to learn as I go. In real world, before we get to the modeling part, we spend a lot of time getting down to the real objective, collecting data, understanding data, cleaning or converting the data to how we want them in our models or analysis. This process takes about 80% of the time. I gain the skills on the jobs.”
In term of quality candidates for analytics roles, in his experience, he emphasizes the importance of soft skills. “In my experience, the data skills can be advanced on the fly, because there are so many free and available resources out there to improve your skills or to figure out a coding problem you encounter; but you need to have good foundation to begin with. Because at any analytics shop, there are at least some main production programs were already written, so if you can read and fix the bugs on the fly, you’re in good shape.”
“The soft skills come much harder. Just as much as I think how important my coding skills and statistical knowledges have been in my career (it is still a constant learning progress), I actually weight more toward communication skills and relationship building in a candidate. In many ways, I see analytics roles in the same way I see a consultant or an internal consultant. From previous roles to my current position at BMO Financial Group, I often engage and collaborate with corporate strategy, finance, marketing, sales, and operations to solve a business problem or deliver a product’s market performance; thus attitude, personality, and communication skills including presentation skills are things I would like candidates and/or seasoned analytics professionals to be aware of.”
Tri has some words of advice to candidates who are seeking to get into the data analytics world. “If you would like to obtain an employment opportunity in analytics field, learn SAS programing language. It is one of the most widely used languages in most major corporations, and the demand curve is trending for the skills. For the new comers and for those who desired to be in the field with no real analytics experience, I recommend you to obtain a SAS certification, because this would provide hiring manager a comfort level of your skills and aptitudes. Personally, I do not care about certifications if you have experience with SAS. Just a degree with no experience and no certifications is not a competitive resume.”
While Tri recommends candidates to start with SAS, but other programing languages such as R and Python have also been the core of advance analytics. “When I was in graduate school in 2011, we used SAS and R as our primary academics languages, I would run conjoint analysis on R and logistic regression on SAS. They are interchangeable in term of statistical modeling capabilities, it was great that we used both. Nowadays it is very common for advanced analytic professionals to mention SAS, R and Python in the same breath. It will depend on what you want to do, but it is not necessary to be really good at all three languages.”
“Ultimately, this is a constant learning profession with new technologies and techniques release fast and often. With the computing power has been exponentially on the rise, now we hear machine learning, deep learning and artificial intelligence. My advice, focus on your core skills, what is needed for the job you were hired for, don’t stretch yourself out with all of the new technologies or different programming languages, you can learn one program forever and never learn it all.”
“If there was someone who was a decent R programmer, but had all of the soft skills, I would put more consideration on this individual over someone who was really good at SAS but did not have the soft skills.”
Tri does not like the proliferation of the term data scientist. Like me, he feels that a lot of the business world’s focus on data science is more of a PR move than something of substance. He views a data scientist as a subject matter expert in analysis, not a role a new graduate can step into. “You cannot be a data scientist until you really know the science of data. That is the hardest part. Typically, someone who has a Ph.D. in quantitative or social science fields, with many years of hardcore modeling and heavy knowledge on C, C++, and Java just to name a few. The definition of a data scientist has recently been loosely defined to fit job descriptions, thus becoming a PR thing. HR puts it out there because it is cool, but that is not really what it is.”
Tri had some parting words of advice for budding analysts who want to succeed. “Once you get technical skills locked down, focus on relationship building, communication skills especially presentation skills. At my current position that is at least 80% of my job.”
What do you think of Tri’s story? Agree with his take on the growth of data scientist job titles? If your story would benefit our readers and you would like to be profiled drop me a note at firstname.lastname@example.org .