10 Tips for Data Analysts

  1. Ask Questions, Nourish Curiosity, and Embrace the Unknown
  2. The business data analytics field is constantly evolving, which makes it exciting. This also means that, to stay competitive, business data analysts need to continually cultivate their skillset. The people who do the best in this field like learning, are self-motivated to try new things, and easily adapt to new environments and technologies. The way to excel and standout is to be curious and relentless in your pursuit of understanding complex problems.
  1. Start Thinking about Everything You See as a Dependent or Independent Variable
  2. The fundamental task business data analyst is to figure out how to make business questions into numbers that can be tested. A crucial consideration is which variables are treated as dependent versus independent. The dependent variable is the measure you are trying to understand. The analysis will determine whether its value is dependent on, or changes in response to, other factors. Those factors are the independent variables.
  1. Start Exploring the Advantages of Continuous versus Discrete Variables
  2. Continuous variables are measurements that can take on an infinite number of values. An example of a continuous variable is a conversion rate for a website, measured as a percentage. Discrete variables have only a limited set of values, measured as "high," "medium," or "low," for example. This decision of continuous versus discrete will have a large impact on the statistical models that are available to use. Discrete variables are easier to understand, but are often less precise. Continuous variables are harder to interpret, but provide more details.
  1. Listen and Contribute (Data analysis projects are almost always collaborations!)
  2. Become an excellent listener, so you are able to incorporate and internalize what other people have to say. But, also, actively participate in group conversations. Both these suggestions become particularly important when the data suggests something that violates the intuitions of other people in the company.
  1. Train your Skepticism Muscles
  2. Whenever there appears to be a really dramatic or surprising effect in the data, most of the time it is due to a bookkeeping or coding mistake. Similarly, if someone comes to you with extreme confidence about a solution to a problem, temper your expectations, because it is likely to be more complicated than they make it out to be.
  1. Seek Details
  2. Real life is messy. It is often necessary to track down details to get to a vantage point where the problem can be completely understood. In general, don't be satisfied with taking data at face value, be curious about it. Dig deeper.
  1. Cherish Precision
  2. Being precise enables you to make progress and rule out bad directions much more quickly than vagueness. Practically, this means making precise predictions, hypotheses, and goals, wherever this is possible.
  1. Best Practices Do Not Equal Common Practices
  2. It is important to understand both the best practices and common practices in the data analytics field. Knowing both common and best will help you determine which practice is most likely to be successful in a given situation.
  1. Expectations Matter!
  2. Almost always, data analysis projects are performed in a team context. The expectations of teammates and stakeholders matter. If the results of your analysis contradict what they expect, it is unlikely to be received well. This will stop your hard work dead in its tracks.
  1. Put Yourself in Other People's Shoes
  2. Just because the field is data-driven does not mean that people are driven solely by the facts. It is important to cultivate the ability to put yourself in the position of another person. Think about what other people are going to feel about your solutions and recommendations. The better you are at anticipating thoughts and feelings, the more efficiently and tactfully you will communicate.

These tips are the personal advice of Professor Jana Schaich Borg who teaches the Data Visualization and Communication with Tableau on Coursera. That course is part of the “Excel to MySQL: Analytic Techniques for Business Specialization” specialization. The specialization is sponsored by Duke University.