Am I a Data Analyst or a Data Scientist?

I should know this one, cold. What’s the difference between data analytics and data science? An acquaintance asked me that basic question recently and my response was half hazard and poorly articulated. I didn’t feel great about that. However, spending an hour googling for the distinction helped me realize I don’t have a good answer, in part, because most people don’t have a good answer. Blog A has a different definition than Blog B; Northeastern University has a different one than UVA; and Wikipedia has a different description from all of the above. Similarly, when I asked my LinkedIn community to differentiate this transportation forecasting problem, the majority was not overwhelming.

So, what is the difference between Data Analytics (Data Analyst, DA) & Data Science (Data Science, DS)? On one hand, you can say that it doesn’t matter and that they both solve problems using data. On the other hand, it matters tremendously if you are trying to figure out organizational design and, more importantly, how to recruit the right talent.

One thing is pretty clear; these groups are tasked with finding patterns in data. While they use a lot of the same tools and techniques (Python, R, regression, classification, clustering, dimensionality reduction, statistics, data visualization, etc.), they do it for different reasons and in different ways. Here are a few initial distinctions that come to mind:

• DA’s answer questions by finding patterns in data. DS’s generate questions by finding patterns in data.?
• DA’s thrive with speed. DS’s thrive with exploration.
• DA’s make you more certain. DS’s make you less certain.
• DA’s toolbox has breadth. DS’s toolbox has depth.
• DA’s sit close to the decision makers. DS’s sit close to the DA’s.
• DA’s describe while DS’s predict. DA’s & DS’s work together to prescribe.

All of these distinctions probably take a very different form depending upon A) the size of the company B) the industry and C) the time frame.

• Size of company. Relatively larger companies probably separate these groups in an effort to realize economies of scale, but unfortunately this almost certainly increases coordination costs and reduces their effectiveness to prescribe solutions.
• Industry. Digital companies have relatively more & cleaner data. This means the time needed to put data together is relatively less and they can probably spend more time on exploration & algorithms.
• Time frame. Less mature data teams, I think, will have significantly less distinction between the analysts and scientists. Longer standing data organizations that have solved for rising coordination costs can have more distinction but maintain a healthy level in effectiveness to prescribe solutions.

What about the pay differential? On payscale.com, the average salaries reported show a >$30k pay difference ($62k versus $97k). On the surface, that’s big enough for me to accept that there is a gap (despite uncertainty in the figures). My experience tells me that the differential of this size is probably unwarranted, but exists because hiring teams make salary decisions about individuals rather than teams. If your goal is to find patterns in data and make decisions with that information, it feels like a great data scientist needs a great data analyst and vice versa.?

All of these features make classifying an individual as a data analyst or data scientist a rather complex, but not intractable, task. Similarly, it makes defining the difference between Data Analytics and Data Science harder than you might think on the surface. I’m feeling better about my response to “what is the difference between Data Analytics and Data Science?” At the same time, I think, write, and publish this publicly so I can upgrade my perspective. Where would you put suggestions if you were editing this article?