Data science has been hailed as a hot career path for some time, with Harvard Business Review going so far as to call it “the sexiest job of the 21st century.

This and other headlines (see Government Technology’s “All Governments Need a Data-Driven ‘Spock’ on the Bridge, Expert Argues” and VentureBeat’s “An AI god will emerge by 2042 and write its own bible. Will you worship it?”) suggest a dynamic field in which practitioners have a laser focus on data and the tools required to transform it into action.

The reality is much more nuanced. In an effort to understand the typical day of a data scientist Matthew Mayo, the editor of KDnuggets, put the question to his LinkedIn connections and documented their overwhelming response in this article. A common theme was that there is no such thing as a “typical day” — not unexpected given the varied nature of data.

What may surprise you, however, is how little time many data scientists actually spend interacting with data.

This finding is echoed in another HBR article, which noted that communication abilities should be a requisite skill for all those considering a career in data science. The author spoke with numerous data scientists for the piece and was told, “the key skills…are not the abilities to build and use deep-learning infrastructures. Instead they are the abilities to learn on the fly and communicate well in order to answer business questions, explaining complex results to non-technical stakeholders.”

It’s certainly true that data scientists must be able to communicate sophisticated concepts to other areas of the business in order to make data more actionable. However, it’s important that the pendulum not swing too far in the other direction. It’s still critical that data scientists have ample time to explore and analyze data, and this is where the industry is currently falling short.

It’s been reported that most data scientists spend 80% of their time finding, cleaning and organizing data, leaving just 20% to devote to more strategic initiatives. As the volume and variety of enterprise data increases with IoT, AI and machine learning, it’s increasingly important that we try to invert these figures.

The right tools and technologies can help significantly. By employing a single platform to integrate, cleanse and analyze data, companies can build an infrastructure that allows for re-use and standardization, freeing up data scientists from some of the job’s more mundane aspects.

Of course, that’s not to suggest that it should all be smooth sailing. By the very nature of working so closely with data — and all the variables it contains — a key component of data science is embracing the unknown. As Eric Weber, a former LinkedIn senior data scientist, said in the KDnuggets article, “An important part of each day is failure….if you are not failing, you are not learning.”

By giving data scientists more time to analyze, experiment and learn from the inevitable mistakes, companies can help their enterprise ultimately become more data-driven.