Career Advice to Data Scientists – Go Make More Money

Data Scientist should offer the enterprise more than the ability (and cost) of doing analysis, but behave as an executive with expertise in analysis and help lead the enterprise on decisions, investments, and operations.

Career Advice to Data Scientists – Go Make More Money!

Each fall, I teach the Analytics Leadership class in the Northwestern Master of Science in Analytics program. We are proud to have very talented students in our cohort each and every year. The graduates will most likely take jobs as data scientists in firms or organizations that expect to get more from their data. As the role and even title of a data scientist has been around for only a few years, it is important to examine what a data scientist should do in mapping out his/her career.

  1. Define Yourself: Are you a data scientist first or a marketing or operations executive with a specialization in data science? The answers are different and suggest a different self-orientation and image in the company. If you are strictly a data scientist, then a move upward, into a role in marketing or operations might be viewed by yourself or others as nearly impossible. It does suggest that lateral moves from one data science job to another are great options in your future. If you are a marketing executive or operations executive, specializing in data science, the language even suggests a more likely path to the Chief Marketing Officer or Chief Operating Officer and beyond. It also means you participate in different types of problems and help the enterprise beyond just solving complex data problems. You participate in guiding the firm through decisions and investments. Embrace the opportunity to do more than analysis; it will lead to more career opportunities, too!
  2. Focus on Revenue: Data science offers many value propositions to the enterprise, but at the end of the day, these values either result in more money made (increasing revenue) or less money spent (reducing costs) and sometimes the blessing of both (increasing margin). (I am omitting special types of problems that create value through tax loopholes and other irregular impacts from regulation). In principle, money saved is money made, and dollars saved are as good as dollars made. However, in practice, costs may not be reduced as much as revenue can grow. Indeed, costs can often be reduced by simple heuristics or rules and brutal policies such as shutting down an enterprise entirely. My point is that saving money for a firm or client is inherently limited. Efficiency can be improved, but rarely repeatedly. Focus instead on growing new revenue for the firm. Go Make More Money (for you and the firm)!
  3. Embrace Easy Wins: A personal trait of many highly intellectual people and technical people is that “hard problems” are viewed as more valuable or admirable. Indeed, universities are full of such people. Solving “hard problems” offers rich intellectual satisfaction. Data scientists may often fall into this trap. “I solved a really hard problem – yeah for me!” Yes, but did it really help the enterprise? What was the value? Work on problems that offer high value to the enterprise. Often these are more attainable than solving hard ones. When working on a project, ask, how does this help the enterprise? Is this problem one worth solving with complex models or general heuristics. Be cautious of really hard problems and unsolved ones – these may be unsolved for good reason – solving them may be of limited value. Hard problems often do not have high value to the enterprise, unless they have enormous scale.
  4. Talk About Data Science: When you get to your first job as a data scientist, it is likely that you will find few peers with the same title. Hopefully, your manager will have a good handle on what to expect of a data scientist, but most likely the rest of the firm will not. Organize “lunch and learn” meetings. Invite your co-workers to an informal meeting at which you present a contemporary topic or important part of data science. Explain a cool data visualization to your colleagues. Present a recent piece of company analysis. Show how data science improved the firm! It will grow your personal network enormously. And most importantly, show how the data science made a financial impact!
  5. Step Away From the Screen: If your goal is to participate in leading and transforming an organization, it will require more than writing code and doing analysis. Those are very important tasks, and your expertise leads to many great opportunities to advance the firm in those areas. Leverage those advantages, but look to other opportunities, too! As your career advances, you will have the opportunity to not just lead analysis but also lead where a firm invests and how it operates. If you work for a company that manufactures goods, go visit the factory Learn how things get done. Learn about the processes that you are modeling. If your firm treats or serves people, learn how service is delivered and received. Talk to employees and customers. In short, really learn the business! It will make your data science better and make you a better executive!

There are currently many great careers openings for data scientists. Nearly every industry has invested in gathering data and the data scientists that help with the data discovery process. However, all firms look to remove costs, too, especially human costs. Data science should be a focus for many firms for quite some time, and that should suggest strong job prospects for data scientists. However, we must acknowledge an important caveat. We have data scientists to solve business problems and to add value to the enterprise. Just as other highly technical professions like paralegals, medical technicians, drafters, journalists, photographers, engineers, and even, now, professors are being threatened by digital models and automation, we should expect new business models to arise that reduce the cost to the enterprise of acquiring the benefits of data science. This will, of course, involve outsourcing. Today much of data science is consumed with the time-consuming tasks of preparing and organizing data. As software tools and algorithms become more advanced, more of the data organization and data preparation can be done without the involvement or cost of a person. These advances are good news for businesses, because they can reduce the cost of doing good data science. It might suggest that data scientists can spend less time on the tedium. Good news! However, caution is warranted.

I recall that the work I did early in my career included a great deal of time spent in setting up data fields and matrices in order to run a simple regression. Even more time was spent on the final graphics. My time was a cost to the enterprise. My benefit to the enterprise was the output. With a host of widely available (and even free) tools, the same tasks (output) can be completed in much less time (at lower costs) today. The graphics are easier to make and look far better, too! Firms look at the work as a cost and work to reduce that cost. The same work with less cost is always preferred. Kaggle is an example of getting good data science done for a bargain. Data science programs have popped up all over the world. More graduates will be available in coming years. To me, this suggests that data scientists cannot be stagnate in their roles, but rather should develop skills to advance in the enterprise. Get close to the business and learn more about how it operates!

My advice is that you should offer the enterprise more than the ability (and cost) of doing analysis. You should position yourself as an executive with expertise in analysis in order lead the enterprise on decisions, investments and operations. Be that expert executive!