Skip to main content

Over the last several years, across all industries, a digital transformation has led to an explosion in the volume and complexity of data sources, and a corresponding expansion of the methods needed to analyse them.  In the pharmaceutical industry in particular, in recent years, we have all seen an increasing interest in the use of real-world data that has led to a demand for advanced analytical skills among statisticians and programmers. With this in mind, there has understandably been a lot of discussion at industry conferences and in trade publications about the need for leaders to develop new technical skills within their data science teams.  However, while this is obviously important, it should not be the only area of focus. In my opinion, soft skills development and the creation of a supportive environment through effective line management and leadership actually have just as critical a role to play in creating a winning data science team.

Outsourcing has impacted our roles

During my career, I have seen a noticeable shift in the expectations of programming and statistical roles.  I believe that this shift comes partly from the increased importance of outsourcing in delivering projects. These changes have affected both sponsor and CRO roles. On the sponsor side, statisticians and programmers alike working with CRO partners to deliver on projects have seen their functions start to incorporate more vendor oversight as well as more traditional technical statistical or programming work.  

CRO programmers and statisticians have also seen a shift.  Earlier in my career, CROs were seen as an ‘overflow’ when a sponsor had an internal resource deficit. Nowadays, outsourcing is more strategic, particularly with respect to resourcing and budget control. The CRO tends to be a partner to the sponsor, with the main role of the sponsor being oversight.  This alteration in emphasis in both environments means our next generation of statisticians, programmers and data scientists will need to have a more rounded experience that encompasses soft communication, influencing and project management skills. In particular project management is becoming an integral part of any analytical professional’s role – and so developing these skills in teams should be given as much importance as building robust technical knowledge.

Setting up your team for success

Team leaders working in both CROs and sponsors have an essential part to play in creating a supportive environment that allows their employees to thrive and set their whole data science team up for success. Here are some of the success factors I have observed:

  1. Distinguish between project management and line management.

    In my experience, there is a distinction between the skills needed for line management and those required for project management.  If this distinction is overlooked, it can be detrimental on all sides. One of the first steps in setting up your team for success is to recognise the different requirements for project management and line management roles and ensure that you create career pathways and learning and development frameworks for your statisticians and programmers that acknowledges this variance. First and foremost- put the right people into the right roles that match their skills and develop them accordingly.

  2. Develop communication channels.

    Line managers need to build open and honest relationships with their direct reports to foster communication and allow the team to thrive.  Creating proper communication channels is not always straightforward, but it is essential so that your team members know they can ask for help when they need it.  In my experience, statisticians and programmers can tend to be reserved about opening up if there are difficulties, so managers should be well-attuned to the communication styles of their team members and observant of any warning signs of problems.  Ultimately empathy is at the core of effective line management and is essential for building a positive growth culture.

  3. Balance learning and development with acknowledgement of experience.

    Leaders of successful data science teams should focus on training for their direct reports that develop their soft skills, as well as their technical skills.  Opportunities to attend and present at industry conferences, or contribute to research publications can all help teams to gain more rounded experience. I’ve noticed that fast track development schemes have also become prevalent within many CROs. These are useful to help develop technical and project management skills for talented, ambitious early-career programmers and statisticians.  However, when thinking about team structure and assigning people to line management positions, let’s remember that team coaching and mentoring is as much of an art as a science. Success as a line manager often depends on acting in situations based on the experience you have built up over time, and dealing with your team instinctively.

  4.  Recognise that effective management will increase in importance as the workplace evolves.

    Data science team dynamics will continue to develop in the future as the world and our expectations of the workplace change. Leaders and managers will need to adapt in line with this evolution.  One key trend which shows no sign of slowing is the focus on work-life balance and wellbeing from all generations. As a manager, you can expect that your team members will expect to have a healthy personal life as well as a fulfilling work life.  Another revolution is happening online- platforms like Glassdoor have provided greater visibility into workplace culture and empowered prospective candidates with greater insight when choosing an employer. So developing a positive culture in which your team can thrive, and which they are prepared to advocate for, will also allow you to continue to attract the best and brightest data science talent.

For busy managers and leaders working on fast-paced projects where delivery deadlines can be tight, and priorities compete, it is not always easy to spare sufficient focus on engaging employees.  However, it is an effort well worth making for the benefit of the employees themselves, the organisation and its customers. I have found personal inspiration from Margaret Heffernan’s book, ‘Beyond Measure: The Big Impact of Small Changes’, in which she argues that it can be the smallest, incremental changes that can have the most significant impact on the success of a team or an organisation.  As leaders managing data science teams, let’s take a moment every day to think of the small acts we can implement that could enhance our team’s skills and nurture the culture.

Watch Diana Stuart’s webinar recording ‘The Generation Game – the X, Y and Z of Line Managing Clinical Data Scientists’

Diana Stuart