Data & Analytics Strategy
Attracting Analytics & Data Science Talent
Demand for data science and advanced analytics roles and skills is increasing, as organizations are looking to extract the most value of their data. According to LinkedIn’s Workforce Report published in August 2018, “demand for data scientists is off the charts” with over 150,000 open roles for positions requiring data science skills. Reflecting this increasing demand over the past few years, the number of job seekers in the space has been expanding as well. Emerging programs ranging from PhDs to week-long bootcamps promise to provide students with the skills necessary to succeed as a data scientist. Searching through the growing pool of candidates for top talent in the data science and analytics space is challenging.
To help attract quality candidates in the data science industry, it’s important to understand what these individuals are typically looking for in a job, how you can best adjust your recruitment processes to communicate the benefits of your analytics organization, and methods of improving the experiences candidates have with your organization throughout the job evaluation process.
In developing this report, we polled and interviewed 45 members of the data science community across the U.S. to help inform our recommended strategy. Topics included the importance of position attributes when evaluating a new job opportunity, paths to overcoming deficiencies in important attributes, and thoughts on job descriptions within the field. While not all findings will be unique to the data science industry, they’re valuable to consider when developing your internal recruitment program.
What’s Most Important to Data Scientists?
Before diving into a recruiting strategy, it’s critical to understand the motivating factors behind data scientists in career changes. The following elements, in order of priority, were deemed most important when evaluating job opportunities.
- Meaning – The ability to make a difference within the organization was rated the most important element when looking for a new job. Candidates want to see that they can create actual value for the company and that the company recognizes the potential value of the data science organization. Not every project will result in groundbreaking findings, but data scientists thrive in an environment where their analysis and recommendations are helping to drive real change within the organization.
- Compensation – Not surprisingly, compensation was rated one of the most important factors when evaluating a new job opportunity. This attribute includes a combination of salary and all other benefits with a monetary component, such as vacation time and health insurance.
- Opportunity – There are two main attributes to advancement and opportunity that are important to candidates. First, opportunity within the company through an established path for career progression is important. Second, if your organization lacks a clearly defined career path, more informal advancement opportunities are also desirable, such as continuing education, technical training programs, or leadership development.
- Challenge – With the significant growth of available data science positions, there’s a concern among candidates about the need for actual data science skills. These candidates are looking to solve complex business or technical problems and will expect the role to provide challenging use cases where they can leverage their education and experience. If your existing data and analytics environment is more immature, consider highlighting future opportunities and the chance to help evolve the data science function.
- Innovation – The freedom to explore and solve problems, along with the flexibility to use various technical solutions and platforms is important to data science candidates. Given the complexity of some data science challenges, candidates want to avoid strict limitations on technology or process guidelines that could prevent them from effectively providing value.
- Team Fit – In parallel to career advancement opportunities, candidates are looking for high-performing, motivated teams from which they can learn and grow. If the team is more junior, the availability of mentors throughout the organization is also a positive point. Since these teams tend to be more collaborative, candidates also stressed the need for friendly, positive coworkers and supervisors.
- Culture – Alignment of personal values with company culture and mission is another important factor. Beyond the internal corporate culture that dictates day-to-day operations, candidates are looking to make a difference in the community at large by improving lives of people outside organization and bringing about positive change in the world.
How to Overcome Limitations
In a perfect situation, your company would have a way to address all elements that are important to a data science candidate. Realistically, your organization may find themselves lacking in one or more areas, and will need to make extra effort to highlight other strengths and benefits when pursuing top talent. There’s no need to turn away an ideal candidate if your salary isn’t competitive with coastal opportunities, your industry isn’t the most glamorous, your location isn’t desirable, or your business strategy is still in development. Instead, you can overcome these difficulties by focusing on what you do bring to the table.
In our survey of the data science community, we asked what a company or organization can do to overcome any perceived deficiencies in their job opportunities or their company. Three main points topped the list: communication, freedom and flexibility, and career development.
Our survey showed that communication was the #1 way for companies to overcome any deficiency they may have. Communication is a broad topic, so what are candidates actually looking for?
To Be Heard – Data scientists want to contribute to something bigger and make a difference. They have ideas and want to work for a company that will provide them the opportunity to voice those thoughts and opinions. They want to be heard. This could come in the form of established procedures like including data scientists earlier in the process during a data and analytics project, creating formal employee surveys, or everyday interactions in meetings. You can share some of these aspects with candidates or ask current employees to share their experiences at a casual lunch in between interviews. If your organization excels in this area, allow your employees to share how refreshing it is to work with leaders who value their opinions.
Current State & Future Goals – It doesn’t take long for an employee to see what the actual state of the union is at your company. Candidates will also form impressions of the environment during interviews and company visits. Be clear about your current state in terms of data, processes, and actions you are taking to improve your situation. Be realistic with the timelines that tie to future goals. Communicating with honesty is incredibly important. It doesn’t benefit any party if the best data scientist on the market is hired, only to uncover after they arrive that the company isn’t ready to utilize their talents. Look for talent that matches your maturity level and actual needs and don’t be afraid to communicate those needs as you look for the perfect candidate.
Providing Freedom & Flexibility is Key
Freedom to Explore – Employees are looking for opportunities to have freedom within their position. Allow data scientists time to explore their own solutions or dive-into helping with another project. Encourage innovation, not just with encouraging words, but with the time, resources, and patience it takes to really support initiatives and projects. Allow experts, not just management, the power to make decisions. Don’t underestimate the power of providing freedom to your employees.
Flexibility in Work Location – Employees are looking for options to work from any location they choose. This may be the ability to work from home a couple days a week, or to flex their schedule to help pick up/drop off kids and then work later in the evening. Some candidates may be interested in working for your organization but don’t want to relocate to your city. If this is not a standard practice in your company, be willing to ask questions to see what it would take to make that type of arrangement work. If you are interested in pursuing remote work options, understand best practices on what it takes to keep these employees engaged.
Career Development Includes Transparency & Support
Transparency – Be open and clear with the career advancement opportunities that are or are not available in your company. It doesn’t help to pretend there is a lot of upward mobility if there isn’t. If you have the ability to create a technical career ladder or reward employees through salary increases or promotions, highlight this benefit and share what opportunities exist. If a technical career path doesn’t exist and promotion opportunities are limited, highlight the support, leadership, and development opportunities that are available.
Support – Companies that show dedication to an employee’s development are very attractive. Providing personalized training plans, a coach, the ability to attend conferences, or financial support for continuing education can go a long way. Employees that want to develop leadership skills will be drawn to opportunities to lead projects, mentor others, plan events for the team, and present to senior team members. Employees will develop a sense of loyalty and commitment to their supervisor and the organization if there is demonstrated interest in helping others succeed. Beyond the managers, having co-workers that volunteer their time to mentor others, to give advice, and support each other can make all the difference.
Creating the Best Experience with your Company
Understanding what data scientists are looking for in a job opportunity and knowing what your organization can or cannot offer is the right place to start in developing a recruiting plan. However, it’s equally important to ensure the candidate has a positive experience as they interact with your company, so you don’t lose them right from the start.
When creating a job description, you’re looking to attract, not scare prospective candidates. It should describe what the job entails, what benefits come along with the position, what skills are required, and provide an accurate representation of the work that will be done. It’s helpful to include the business area and strategy in addition to the job duties to really describe the position. A prospective candidate will look beyond the title to see the role’s position within the organization, the scope, and the actual responsibilities of the role.
Be aware of the experience candidates will go through when applying at your company. As many HR departments have adopted software to aid in reviewing applications, the process may have become unnecessarily cumbersome and difficult to navigate. There are often many hand-offs in the process and it’s easy to make mistakes. A survey done in 2016 showed that only 20% of candidates received an e-mail notification that they were not being considered. Take ownership of communication with the candidate throughout the process. It can make all the difference.
At some point in the hiring process, it’s also very beneficial to have someone from your company share more information about the benefits of working for the company. This includes anything from health benefits to onsite fitness facilities to the flexible work environment that the company offers. If diversity is important, review the different ways diversity is encouraged and supported. Remember, impressions are made based on the efficiency, clarity, and integrity of the entire hiring process, not just the interview.
Interviewing Best Practices
It’s important to design an interview and assessment process that is best suited for your ideal data science candidates, accounting for desired skillset and experience levels.
Avoid hyper-focused interviews. If interview questions are too focused on a narrow skill set, it’s easy to miss out on a candidate that could fill another role on the data team. It’s helpful to provide a test that can be given to candidates regardless of education, with various expected response types based on experience level and area of expertise.
Look beyond algorithms and analysis. A solid functional understanding of data practices and principles is necessary for any member of the data team, but other less quantifiable skills are just as important. Curiosity, communication skills, and overall team fit are not evaluated by applying an algorithm to a statistical problem. Task the candidate with presenting their results to your team. This is extremely important and helps weed out candidates who put together an impressive written report but fail to effectively communicate their results. Not only does this step help your organization gauge the communication skills of a candidate, but it allows you to evaluate cultural fit.
Don’t create unnecessarily stressful environments. Don’t whiteboard test candidates in real-time. This practice adds unnecessary stress to an environment that’s inherently high stress and not particularly relevant to real-world situations. Memorization of algorithms, a necessity for whiteboard tests, is less important than an understanding of data, analytics, and statistical principles.
While it’s helpful for the candidate to meet with existing team members to assess cultural fit, avoid a gauntlet of all-day interviews. You can artificially narrow your candidate pool with the requirement of an all-day in-person interview, which isn’t feasible for some candidates.
Stop looking for a unicorn. Don’t expect to hire one person that excels at everything from data munging, engineering, analysis, visualization, application development, and executive presentations on findings from data analysis. The person responsible for managing your data team should take your test and determine the skill levels and responses appropriate for your open positions. Look for a well-rounded team that provides unique perspectives.
Note: This article was written in collaboration with Jenny Schmidt and originally published through the International Institute for Analytics: https://portal.iianalytics.com/research•