Beneath the mountains of data now available to companies are tangible insights waiting to be unearthed. Business leaders are realizing the value this can bring to their companies, as evidenced by the incredible rate at which they are hiring data teams.
Yet the job description of data scientists is changing. The rise of artificial intelligence, machine learning and automated data collection means the nature of their work is shifting at a rapid pace. Time-intensive tasks, including data collection, are becoming less primary as they become automated, giving way to more strategic tasks.
According to Kaggle, data scientists currently spend up to 80% of their time on data collection. Because data is often incomplete, outdated or stored in the wrong format, teams frequently have to update, clean or reformat data sets before they can be used.
Today, tools can automate many of these processes. Gartner estimates that by 2020 more than 40% of tasks performed by data scientists will be automated. This frees data scientists up to add even more value to their organizations, spending more time on insights and analysis.
This transition, however, means data scientists need new skills for success.
The old job description for data scientists put a heavier emphasis on data collection work; the new job description emphasizes strategic application of that data via technology, programming, and communication skills.
Artificial intelligence and machine learning, in particular, offer data teams new ways to explore information for critical insights. Though only a handful of data scientists are now proficient in machine learning techniques, those that begin to practice and advance these skills can dig deeper into their data for advanced insights. What they find may even surprise themselves as these technologies uncover results that may not be readily apparent or might go against the grain of accepted wisdom.
With greater time for analysis and strategy, thanks to automation of tedious collection tasks, data scientists can more quickly turn their attention to creating insights for executives. This is why tomorrow’s data science leaders also must focus on soft skills.
The ability to communicate clear, evidence-based recommendations should not be overlooked. Before business leaders can take action, they need to understand the how’s and why’s that are driving future performance projections. Data scientists can make these relationships clear. However, their insights need to be packaged so that they can be easily understood by less-technical executives. When creating reports and devising recommendations, teams should use graphs and visual projections to communicate the full value of their data-led observations, focusing on outcomes and results over the processes used to attain those insights.
Data scientists also need to see the big picture of business goals and strategies. Regular communication with business leaders can help data scientists provide context to their numbers. With knowledge of external forces, industry trends and company-specific initiatives, data scientists can ensure their insights align with corporate goals. For example, if sales are struggling in one market but not another, data scientists should seek to learn how execution in each market varied and what forces might be at play influencing behavior.
Armed with this knowledge, data scientists can look ahead for insights into the future. They can begin to answer more complex questions like what trends will have the biggest impact on demand next year, where should the company invest its ad dollars and how the market might react to a new product.
In this way, data scientists of the future will spend more time on tasks that drive value for their companies as automation facilitates more strategic, forward-looking work.