What makes a great data scientist

Share on linkedin
Share on facebook
Share on twitter

What makes a great data scientist?

Your organisation has decided to start your own data science department and all that remains is to find suitable talent. Easy, right? Not really. Successfully hiring a data scientist can be a little daunting especially if you don’t have an in-house analytics department.

Not only is it hard to find good talent, but McKinsey estimates that for every data scientist, organisations will need ten data-savvy managers with the skills and understanding to make strategic business decisions based on the data analysis generated. This means that even if you hire a data scientist, there is huge pressure on them to be able to communicate their findings to your organisation in such a way that drives the most value from your data.

We have compiled a free list of our interview questions for potential data scientist. If you would like to receive this free list, fill out the form below and we will send it to you.

To further help you hire a data scientist who can do this, we’ve put together a list of what differentiates a good data scientist from a great one.

Let’s start with the minimum requirements. Obviously an up-to-date knowledge of the tools and techniques is essential. At a basic level, data scientists need to be strong in data analysis, statistics and coding (SQL and R or Python). A background in predictive modelling is highly beneficial.

However, knowing the technical aspects is not what makes a data scientist great. Imagine how ineffective a data scientist would be if they knew all the latest tech but spent their time tinkering with the latest and greatest methods instead of linking the most appropriate method to the business requirements. It is not enough to be interested in coding yet have no desire to solve problems or generate value for the business.

So what are the skills and characteristics that make the magic happen? Here’s a list of our top favourites:

1. Statistical Thinking – knowing how to create algorithms is one thing but knowing when and how to apply them in a stable and reproducible fashion can be an art form in itself. An intuitive understanding of statistics in combination with a healthy dose of scepticism enables data scientists to determine when results seem suspect. This “sixth sense” acts as a protection system that sparks renewed questioning so that potential blunders become game-changing insight.

2. Technical Flexibility – technology is changing rapidly and being able to learn and quickly implement learnings are vital skills. While experience is important the “hacker’s spirit” is equally important.

3. Ultimate Communicator – a great data scientist needs to be able to take algorithmic output and clearly communicate it to interested parties of various backgrounds. To do this they need to be able to tell a good story, create graphs where the bottom line stands out and be intuitive listeners so that they can adapt their message if their audience is not following them. Not only do they value how well others understand them, but they also understand the importance of understanding the business problem fully and do not shy away from asking questions.

4. Good Judgement – this comes hand-in-hand with an understanding of the value of time. A great data scientist knows that an imperfect solution delivered on time is better than a perfect solution that is late. They understand that solutions will need to be reworked as new insights or data come to light and they avoid analysis paralysis.

5. Curious – curiosity triggers the need to explore further than what is required. It results in trying something new, asking for access to more data and being quick to question their own assumptions. They are obsessed with solving problems and not just with new tools.

6. Show Initiative – initiative combined with curiosity enables a great data scientist to take advantage of change opportunities that arise.

7. Creative – great data scientists struggle to accept that they can’t do something. They feel a compulsion to overcome the seemingly impossible. While their initial response might be “no”, their close second is “wait a second, let me think about it”.

8. Persistence – the world of data analytics is messy at best. Data can be missing or incorrect, technologies are constantly changing and business needs require fast, accurate results. These challenges and unforeseen road blocks can be frustrating but a persistent data scientist has a taste for tackling challenges and rising above them.

If your organisation is not yet ready to hire its own data scientists but is wondering what can be done in-house without one, click here to read this helpful article from Gartner. Alternatively, consider outsourcing your data science needs to a company like Incline that already has the infrastructure and skillsets that you need to access.