To say that data science is a growing field would be an understatement. In Linkedin’s 2020 Survey, data scientist was the third most sought-after profession in the United States and number seven in the UK. If the trend continues, data science’s market will see a 26% growth in the next 5 years.
Why has data science become such a prevalent field? The simple answer is data culture. We have cultivated the idea that information and data can provide accurate answers and guide our decision-making process. This has led to a bigger market for specialists who can gather it and analyze it.
In addition, other factors driving the growth of data science are:
- Technology is getting better: With more powerful computers and better tools at our disposal, gathering and processing huge amounts of data has never been easier. Yet, machine learning, artificial intelligence, and cloud computing all require the guiding hand of a good data scientist.
- More data: The amount of data being produced is nothing short of overwhelming. Some suggest that 2MB of data is created every second by every person in the world. The scale is simply unimaginable. As more fields are adopting data gathering, more data scientists are needed to process it.
- Easier access to big data: With the increase in data and the development of better infrastructures, data sharing has grown exponentially in the last few years. Even small-scale businesses can afford to invest in big data and the workforce needed to put it to good use.
Suffice to say, any data scientist worth a dime has to have a very firm grasp of Python and R. Both programming languages are well known in the community and, if the trend continues, Julia might become the next great programming language in data science.
But doing data science is more than knowing how to code and having a firm grasp of math and statistics. What other skills can a data scientist develop to give them an edge? What should you look for as a secondary skillset in your future employee?
Yes, a data scientist should have a good grasp of databases, especially relational databases, as they are the most popular and widely used technology on the market. They don’t have to have the expertise of a database engineer, but they should know enough about schemas, tables, and queries to understand the fundamentals of storing data.
Creating and managing databases is a science unto itself. Drawing a good entity relational diagram from the get-go can be the difference between a scalable database and a constant headache. Data scientists who understand how SQL databases work tend to be more conscious when manipulating and altering databases, implementing time-saving and resource-efficient strategies.
Service solutions such as SaaS are becoming the norm in the IT industry, and while any company who actively works in the cloud would do well in hiring a cloud engineer, everyone who is involved in the project should understand the basics of working on the cloud – and that includes the data scientist.
Cloud-services like AWS are offering powerful data analytic solutions like Amazon Redshift, a data warehouse specifically developed for data analysis and sharing, and H20.ai, which automates some of the most difficult data science and machine learning workflows.
Aside from specific data analysis tools, data scientists who are going to be working on the cloud need to understand how budgeting works, what are the implications of increasing processing power or memory usage, and how to best optimize their resources.
Data visualization and storytelling
For a long time, data scientists were regarded as some kind of magicians or wise hermits who would lock themselves in their office and come out with an executive summary of their findings. It didn’t matter how they did their magic as long as the results were useful for the decision-makers.
But things have changed, and data scientists are no longer loners with magical powers. Most people who work with data are expected to be able to produce robust results and present them to an audience, and that’s where data visualization and storytelling kick in.
For people with a background in math, numbers on a table can tell a whole story. However, they are the exception rather than the norm, and most of us need another representation of the data to help us make sense of what it can tell us.
Data visualization is the science of creating graphs, tables, and other representations that inform people about data. It has come a long way from the times when pie charts were used for anything and everything.
Choosing how to visualize data is a challenge unto itself. A good data scientist understands the psychology of their targets and chooses how to present data based on that. The same information can be relayed in different ways, depending on who the audience is.
Visualization goes hand in hand with storytelling. A data scientist is a storyteller whose job is to align the findings from empirical data with the ongoing story of the company. That’s why they ask questions like “How can their work help our business achieve its goals and get closer to its vision?” and “How can they translate this information into terminology that’s aligned with the company’s culture?”
As we said in the beginning, data science has grown in popularity due to the surge in information-based decision-making. Investors and owners want certainty, and that’s something that requires more than a good model.
A data scientist can’t live in an ivory tower, interpreting their data with the objectivity and cold logic of a researcher, creating models that might be extremely precise, but rather useless in the real world.
There is little to be gained from knowing that a variable predicts consumer behavior if there is no way to control that variable. From the get-go, having a clear understanding of business strategy helps the data scientist in making informed choices about the nature of their work and in turn providing solutions that are aligned with the expectations of people from outside the field.
Motivation towards learning
Last, but certainly not least, the field is growing at an accelerated rate. Old models are being deprecated in favor of better alternatives. That’s why data scientists owe it to themselves and their coworkers to keep learning about new approaches, tools, and technologies that will help make a better analysis.
The market is overflowing with master’s degrees and data science specializations of varying quality. Universities are opening their door to scientists from all walks of life who want to start a career as data scientists.
There are thousands of books published on the subject, as well as summits, events, web pages, communities, and social networks. There is little for a data scientist to not be up to date with the latest research in the field.
Wherever you want to hire data science for your business or you are thinking about outsourcing, always try to look beyond the core skills in a resume. Secondary skills are often the extra value that differentiates between just another team member, and someone who can help you shape your dream.