Data Scientist Flash Survey Results
As part of our ongoing efforts to ascertain what challenges data team leaders are facing going into 2020, we recently sent out a flash survey to our community of data team leaders. Following is a summary of our findings regarding their challenges and plans for hiring and developing data scientists in 2020.
Are Companies Experiencing a Shortage of Data Talent?
In mid-2019 we examined the question, “Is there a data scientist shortage?”. Back then we concluded that, yes, there was still a shortage of data talent. So we asked data leaders if they were still experiencing a shortage of data talent.
38% of respondents indicated that their team or company was experiencing a data talent shortage. An equal number however, responded that they were not experiencing a data talent shortage, with a quarter indicating that they were “unsure”.
In IBM/BurningGlass’s 2017 “The Quantcrunch” study experts found that it took at least 5 days longer to fill data science and analytics roles than for most other job roles. For senior and highly technical roles, the delay was even greater. These delays still seem to exist.
54% of our respondents indicated that the time to fill a data science role was either “too slow” or “painfully slow”.
38% said it was “about right” – so perhaps some progress has been made on this front.
Conclusion: While some progress perhaps has been made on addressing the data talent shortage, data talent is still hard to find for many companies.
What Challenges are Encountered When Recruiting for Data Science Roles?
We asked data team leaders what key challenges they faced when hiring for data science roles. The chart below shows the percentage of respondents indicating which challenges they face.
We were somewhat surprised to see that budget and salary constraints ranked #1 with 45% of respondents experiencing budget issues when hiring data scientists. This is a sure indicator that the market for data scientists is still pretty tight and that qualified candidates are still able to command salary premiums.
Budget constraints might also signify that companies are not allocating enough recruitment resources to finding what is still relatively rare data talent.
These salary constraints make even more sense when we observe that the #2 challenge noted by survey participants is a “lack of qualified applicants”. Clearly this lack of talent contributes to the salary issues in that once again, job candidates to a certain extent have their pick of jobs.
Equally challenging is the fact that that the hiring process for data science roles takes too long. Again, with a lack of qualified candidates and limited hiring budgets, it makes sense that finding a qualified candidate who will accept a budget-friendly salary would take longer.
Conclusion: High data science salaries will continue to present challenges for many companies looking to expand their data teams.
Are Data Leaders Satisfied with the Hiring Process for Data and Tech Roles?
Despite the noted hiring challenges, about 50% of respondents said that they were satisfied with the “process” of hiring data and technical talent in their company. That’s encouraging.
The other 50% were more “neutral” about the process.
What is the Most Difficult Data Science Related Skill to Find?
In light of all the recent hype regarding the importance of soft skills in the data science field, we expected these skills to come out on top for this question. However, good ole technical skills are still the biggest challenge to find.
We were surprised to see “common” data science skills such as statistics and visualization in shortage.
Soft skills did rank second in terms of hard to find skills, but technical skills were twice as hard to find.
The Importance of Diversity in Teams
An encouraging 79% of teams said that they proactively seek to hire gender or ethnically diverse teams.
Roughly 54% believe that standardized skills tests can help to reduce bias in hiring (we agree!).
Learning and Development Investments Planned
55% of the respondents are investing in learning and development for their team.
Online learning is the most popular outlet for investment with 25% of teams using this outlet for L&D. However, conferences, workshops, and tool-specific training are just as important with 21% of teams planning to continue using these.
What Factors Contribute to Data Scientist Career Development and Success?
We asked participants what factors they feel have contributed to their career success in the field of data science.
We were reminded of the intellectual challenge-seeking and curious nature of data scientists by their responses to this question.
Education, in the form of advanced degrees and elearning was the top factor contributing to career success (with internal training not far behind). These two forms of education, to a certain extent, could be considered two opposite ends of the data science education spectrum, reflecting the ongoing boot camp vs. master’s degree debate.
That said, these results may also reflect the fact that the top and most highly qualified data scientists are also the ones who recognize a need to continue learning on the job primarily via eLearning.
Data leaders take note. Mentorships were also very important for the majority of respondents.
Conclusion: Data science is a highly technical and challenging field that attracts curious, life-learners who rely on continued and quality education to remain at the top of their game.
It would appear that the storyline for data science talent remains more or less the same for the time being. Skills and qualified candidates are still hard to find. This prolongs time to hire and puts pressure on hiring budgets. As such salaries remain elevated and companies struggle to pay for talent. The war for data talent continues.
Are you experiencing any of these issues or agree with our respondents? Let us know in the comments!