Reflecting on a Decade of Data Science and Implications for Future Visualization Tools

Data science has exploded over the past decade, reshaping business and preparing the next generation of workers, but its rapid growth has also created ambiguity about how to extract actionable insights from massive data sets.

Motivated by personal reflection and a desire to identify unmet needs for visualization analysis tools, the author reviewed literature and presented findings from a study titled “Passing the Data Baton: A Retrospective Analysis of Data Science Work and Workers”.

The study defines data science as a multidisciplinary field that applies statistical and computational techniques to learn new insights from real‑world data, emphasizing challenges of scale, real data versus simulated data, and the integration of domain expertise.

Data science work is broken down into four high‑level processes—preparation, analysis, deployment, and communication—and fourteen lower‑level tasks, with visualization playing a key role in several of them.

Through a meta‑analysis of twelve studies involving thousands of data scientists, nine distinct data‑science roles were identified, ranging from statistical specialists to ML/AI engineers, highlighting the diversity and fluidity of skill sets across the field.

Understanding these roles and processes informs the design of better visualization and analysis tools; current tools often focus only on model visualization and neglect other critical stages such as data preparation, deployment, and communication, increasing overhead and limiting impact.

“Data science is a multidisciplinary field aimed at learning new insights from real‑world data through the structured application of core statistical and computational techniques.”

By applying this framework, developers can create more targeted, evidence‑based tools that support the full spectrum of data‑science activities and accommodate the varied needs of different practitioner roles.