It’s been a pivotal year for the data science industry. Framed as ‘the sexiest job’ of 2020, data science has proven its applicability to a variety of industries – from healthcare to agriculture.
The constantly maturing market requires cohesive end-to-end data science solutions endowed with diverse expertise. For years at Newfire Global Partners, we’ve been sourcing data, building models, and operationalizing machine learning insights for future-thinking companies. Here are the seven most important things (backed by research) that we’ve learned about data science throughout this unprecedented year:
#1 – Python Is King
A comparative analysis among 65,000 professional developers has shown that Python remains preferable for 66% of developers. It is more loved than Java, R, and C#!
Python’s extensive selection of frameworks and simple syntax makes it easier for programmers to focus on problem-solving, rather than tinkering with complex algorithm testing.
Due to the rapidly growing library ecosystem and collaborative community, Python’s popularity for machine learning and data science is projected to continue through 2021 as well.
#2 – Data Management is Time Consuming
While moving from hype to maturity, data professionals have much work to do before actually delivering actionable insights. And a lot of this work is about ingesting, cleansing, and loading chunks of data.
The fact that data wrangling still takes a lot of time negatively impacts overall job satisfaction as well as productivity. We expect an ‘effectiveness gap’ will emerge so that the industry can start working on a solution.
#3 – A T-Shaped Skill Set Is a Must
A data scientist’s routine requires a very diverse skill set – from math and modeling to visualization best practices and proper DevOps deployment. Having a strong handle on these various components will be a must-have skill for a thriving data scientist. The ability to create across disciplines is as important as fluency in core technology.
Nearly half of the CIOs in a Gartner survey said they were in the market for employees with AI skills, yet 37% of those same respondents found such qualifications hard to recruit. In fact, slowed hiring for AI was cited as the biggest barrier to adoption in MIT Sloan & BCG Henderson Institute study. About 80% of respondents said they lacked the needed skills to manage AI programs.
#4 – Open-Source Is The New Black
As with any development process, data science comes with inherent security challenges. Open-source projects, unlike proprietary software, enable their supporters to fix issues and vulnerabilities quickly. Additionally, a data scientist can use them for experimenting with algorithms without worrying about payments or vendor lock-in.
However, while developers value the speed of innovation coming from open-source solutions, there are security challenges that come along with open-source projects. Data scientists tend to take a more hybrid approach, using both open-source and proprietary software.
#5 – The Value of Data Science Continues to Gain Ground
Demonstrating the effect data science has on an organization’s business outcomes is essential. However, the degree to which data scientists feel they can demonstrate the impact of their work varies. In terms of demonstrating value, data scientists in the IT and financial industries have the highest rates, while those in healthtech say they’re able to do so only one-third of the time.
#6 – Ethics Should Be Taken Into Account
As the impact of data science grows, it affects our day-to-day business approaches, society, and politics, which brings up some complex questions around ethics. Fairness and personal responsibility are considered must-have soft skills for every data scientist, software developer, or general business executive.
Anaconda’s findings show that only 15% of universities are offering ethics training, while business knowledge is also only covered in 15% of data science programs.
As data science becomes increasingly keyed to business outcomes, we suggest investing in training activities geared towards business-related ethical considerations. Ethics is already an essential component of a data scientist’s work and will become even more critical in the coming year.
#7 – Social Impact Is A Big Deal
Given the increasing impact of data science in our lives, data professionals grapple with big questions like inclusiveness, job automatization, and information warfare.
According to Anaconda’s survey, the bias in data modeling and the impacts to individual privacy rated the highest amongst problems to be tackled in data science.
Forward-thinking organizations should treat ethics, fairness, and open-minded culture as strategic risk vectors. In order to affect change, these considerations should be constantly threaded through data science work processes.
Looking ahead
We’re optimistic that the industry is on track for continued growth and that data science will eventually become a strategic business function. Spreadsheets are finally taking a backseat to data visualization and dashboards providing actionable insights. Embedded analytics has simplified data-driven decision making.
However, many organizations still underinvest in the data-first strategic transformation of their culture and processes. As data specialists, we understand that creating a culture that embraces data is a complex process requiring bold moves and new (sometimes out-of-the-box) ideas. We are experts in crafting a strong data foundation within organizations and tackling specific data issues.
Are you interested in learning more about whether our expertise aligns with your 2021 data needs? Check out more about how we strategically partner with companies around data and let us know how we might be able to help.
