Communication and Collaboration Are Key
For data scientists, there’s so much talk about methods and keeping abreast of developments in the field, but the importance of remembering the basics is often underestimated.
Don’t try to use a complex technique just for the sake of the technique, but remain aware of developments in the field by others solving similar problems. Remember, the goal is to answer a question, not to use a model.
Communication has always been important, but as methods get more complex, the ability to abstract away from the methods to construct the big picture is vital for building the credibility of the data system and driving a data-driven culture.
A good way for data scientists to work on their communication skills is to simply practice. They should hold periodic meetings within the team to not only discuss decisions and methods for analysis (building the technical skills and learning from peers) but also discuss the impact of the analysis and extract the important points. This focus on impact and the highlighting of the most pertinent points will translate well to discussions within the larger organization.
After achieving comfort in a supportive team environment, data scientists should share their work and results more broadly, focusing on context and impact. If an appropriate formal session exists, capitalize on it. If not, there are less formal options available. For example, the data science team can work data into a hackathon project, sponsor a demo-days happy hour, or offer lunch-and-learn sessions.
Another approach is to offer classes about the data that exists in the organization. People are curious, so organizations should encourage them to explore data and the reports that might already exist, and empower them further by teaching them the basic skills for working with data on their own. In the best-case scenario, members of other (non-data science) teams who have used the work of the data science team can share how they used that data and analysis to make better, data-driven decisions.
For more articles like this one, go to the 2020 Data Sourcebook
What’s Ahead
The massive rise in data science roles isn’t slowing down anytime soon, according to the “LinkedIn 2018 Emerging Jobs Report.” Therefore, we will only see greater complexity and variation emerge within the role—further compounding the challenge of overcoming the skills gap.
In addition, people continue to have vastly different perceptions about the meaning of the term “data science.” There’s a clear lack of consistent definitions in the space. Job descriptions vary drastically, and the understanding of the problems for which data science can be applied varies as well.
In some companies, a data scientist might contribute to the development codebase. In others, data scientists do specific, in-depth analyses. Then there is the whole range in between. Some companies have begun to differentiate titles, but that hasn’t happened across the board yet.
Methods are also becoming more complex and specialized, which could lead to an increased diversity of skills and experiences within the team. This variety may elevate the team’s output, especially if coupled with strong communication and a lack of ego. Dialogue between team members about approaches, assumptions, and decisions can lead to better versions of analyses. Citizen data scientists can also serve to contribute to the team by adding subject matter expertise and holding the team accountable to high analytical standards.
With so many variables at play, only time will tell how the data science field will develop and evolve throughout the coming years. But in the meantime, organizations can arm themselves for the future of data science by following the above guidelines—building a culture and curiosity around data, emphasizing communication and collaboration, and empowering a new generation of citizen data scientists who can help overcome ever-rising data demands.