How to Choose the Right Data Science Position in 2023Feb 14, 2023
Looking for a job in data science is overwhelming no matter who you are. There are a lot of steps and a lot of interviews, so where do you even start?
For the data science job search, the first step is targeting the right position. This helps you narrow down your search and figure out what skills to prioritize.
There are two main types of data science positions to choose from: analytics-driven and algorithm-driven. In this blog, we will be looking more closely at these and how you can tell which one is right for you.
Finding Your Target Position
There are a lot of positions that fall under the category of data scientist. If you want your preparation and search to be manageable, you need to narrow it down. The type of position you are targeting will determine what the interview process looks like and how you need to prepare.
If we are being comprehensive, we would say that there are four types of data science positions: analytics, algorithms, statistics, and engineering. You can find descriptions of all four in another of my blog posts.
For this blog though, we’re going to focus on analytics and algorithms because these are by far the most popular. Here are some example positions under each type:
- Data Analyst
- Product Analyst
- Data Scientist-Analysis
- Product Data Scientist
- Data Scientist-Algorithm
- Data Scientist-Machine Learning
- Applied Scientist
- Research Scientist
Before we dig into the specifics of these two types, I want to clarify one point. The division between data science roles will be much clearer at larger and more mature companies. At small companies, especially startups, a data scientist might have a more fluid role that would combine the two types.
Now, let’s dive into these two types!
Analytics-driven data scientists investigate data to make recommendations and drive business decisions. While it is a technical role, it also involves a lot of communication to present data and help stakeholders make informed decisions.
So, what are the requirements for this type of role?
Based on analysis of actual job posts, for a basic analytics-driven position candidates are expected to have proficiency in SQL, an understanding of scripting language for data processing and development, and expertise with statistical data analysis and A/B sampling methods. Communication and collaboration skills are also important.
Other skills that are not as strictly required include software development experience and an understanding of machine learning algorithms.
Responsibilities typically include things related to product ideas and making business decisions. The roles also typically imply the need to be familiar with A/B testing.
As you move to more senior analytics-driven positions, the responsibilities remain similar while requirements begin to focus more on having experience alongside expertise in SQL, Python, and R.
To summarize, data scientists in analytics roles gather data and interpret it to help make business decisions. It requires technical know-how and soft skills like communication.
In an algorithm-driven role, instead of working on products and business insights, you’ll be working on developing and improving machine learning models. That’s why these roles are often called Data Scientist - Machine Learning.
You will need strong coding skills to work in these roles because you will sometimes have to work with software engineers. Requirements from real job posts emphasize machine learning and proficiency in scripting languages such as Python.
Responsibilities typically hinge on building, developing, and using statistical and machine learning models.
Therefore, these roles require heavy technical know-how, especially familiarity with machine learning.
Comparing Analytics-Driven and Algorithm-Driven Roles
You’ve probably gathered that algorithm-driven roles are related to machine learning and analytics-driven roles are more related to products and business insights.
That is the biggest difference, but it’s important to highlight some of the other differences.
In terms of what you need to know for interviews, analytics-driven roles require proficiency in SQL while algorithm-driven roles tend to care less about SQL and more about the scripting language. Analytics-driven roles also emphasize soft skills more while algorithm-driven roles are more technically focused.
Another important difference is that the analytics-driven track tends to have more job demand. It is the most common role you will find in the data science job market, and that is something to consider when deciding what position to target.
Which Position Is Right for You?
Now that you know more about each role, how do you choose what’s right for you?
Obviously, your personal interest should play a factor here, but there are some objective things to consider based on your background.
Data Scientist with a Couple of Years of Experience
Data scientists with at least a couple of years of experience who want to switch to a different company or work in a different field are in a good position. All types of data science roles in most tech companies should be viable for you if you can demonstrate relevant industry experience.
Working in a Startup and Want to Advance to Bigger Companies
If you’re working in a startup and want to move to a bigger company, sticking to the same track should give you a very high chance of getting interviews. So if you have been working in the analytics track at the start-up and you apply for an analytics position at a bigger company, you would be an attractive candidate.
If you do want to change tracks, you can still be very competitive as long as you can demonstrate that you have some relevant experience or have been continuing your education with relevant skills.
The bottom line for both of these first types of job searchers is that if you have data science experience, you can usually target either analytics-driven or algorithm-driven roles and have success.
But what if you don’t have any experience in data science?
Let’s say you’re a new grad. The right type of position will depend on your education.
A statistics and business focus would make the analytics track the best choice. If you majored in computer science, however, the algorithm track would probably make more sense.
What about career changers who want to break into data science from a non-technical role or from a non-tech company? My general recommendation for these people is to focus on analytics-driven roles.
This is firstly simply because there are more open positions in the analytics track. Also, for career changers, the skills you have gained in your previous work will likely be more transferable to the analytics track where product and business knowledge will be very helpful.
This does not mean that you cannot aim for an algorithm-driven position, but you will need a strong foundation in coding, and if you don’t already have some knowledge, it could take months or even years to gain.
Once you know what position to target, your job search will have more direction, and you can start taking the next steps to land a data science job.
If you want to dig deeper into these roles, I suggest checking out the longer version of this post here.
And if you want to learn more about those next steps, check out this post on getting interviews.