There is a lot out there about what machine learning, statistics, and programming skills a data scientist needs to know. This post is not about that. This post is about the often overlooked skills that aspiring data scientists forget about when they are trying to break into the data science and analytics field.
Data science and analytics has turned into a knowledge arms race where everyone is worried about learning the next ML algorithm or programming something or other and they kind of forget that the point of their job is to bring valuable and actionable insights to their employer.
Is math, stats, and programming necessary to be successful in data science and analytics? Obviously. These skills are necessary but in no way are they sufficient. Now with that being said, here are three essential skills you’ll need as a data scientist:
- Have good business/product sense
I think what a lot of people forget (especially those coming from an academic background) is that a company is hiring you to be able to use data to help the company’s leaders make decisions and reach their goals (whether that is to grow, make money, further develop a project, etc.).
In most cases, a data science project starts with some initial question or problem and the data scientist has to think through a few things:
- What data is available, how do I need to clean/format/process it?
- How will I approach my analysis? What statistical, mathematical, and technological knowledge should I leverage?
- What is the point of my analysis? What’s the use-case?
In my experience, point #3 gets lost in the data science project process way more than it should. Every graph, chart, table, and result you present from your analysis needs to be actionable and insightful. An executive or manager should be able to look at your results and not have to guess how they are supposed to feel about it. A good data scientist understands the goals, products, and initiatives of their company–and is able to frame their analysis to help the company understand if they are reaching their goals. To be effective at your job, you need to make sure the decision makers understand why your analysis is important and what it means for the business. An incredibly important skill is being able to craft a narrative that gives concrete solutions to the business problem.
- Be able to defend a technical model/algorithm to a nontechnical audience
The vast majority of a data scientist’s work is inaccessible to a non-technical executive or company decision maker. Throughout the lifetime of a data science project, you make decisions during data cleaning, assumptions during data exploration, choices during modeling… the list goes on and on. If you’re preparing a presentation to leadership about your analysis, you need to think through what questions and concerns might arise so that you can strike a balance between answering their questions without getting too much into the weeds about your work (but make sure all your assumptions and decisions are written down and recorded so you can defend your work to your technical team members!).
A time when I had to deal with this was when I was putting together a model to predict how many subscribers would churn if certain premium features were no longer available to subscribers. This project had huge business ramifications–on the finance, marketing, and product side. Because of this, there were a lot of questions from the executive team when I presented my work. However, the tricky part was walking through certain decisions or assumptions I made without going into insane technical detail about why I did something a certain way. Give examples–and craft a narrative that is easy to follow. Your work means nothing if you are unable to get buy-in from the leadership team.
- Be able to explain poor results/findings with tact and thoughtfulness
This happens quite a bit. As a data scientist, you will often end up analyzing and evaluating a new product feature, initiative, marketing strategy, etc. that ends up failing. You need to be honest, but giving “bad” news is definitely a challenge. You need to understand that a lot of time and effort goes into a new feature, initiative, or strategy, and being mindful of that is important. Your job isn’t to drop data bombs and call out people and teams–it’s to be a resource and to help the company reach its goals.
Interested in a step-by-step guide for how to get into the data science and analytics field? Check out my guide.