How to Be a Great Data Team Manager

Scaling AI Nancy Koleva

Data team managers today are tasked with the difficult but crucial job of leading and empowering data teams to deliver insights and drive data value across the organization. This, of course, is easier said than done: in order to excel at their job, data team managers need to master a variety of hard and soft skills, tackle complex relationships with different stakeholders and take into account a number of key considerations for the overall success of data and AI projects

1. Understand the Business and Its Stakeholders

In order to build and manage great data teams, you and your team need to have a clear understanding of what the business goal behind the project is. Anchoring your team’s efforts in the broader organizational and business objectives is one of the most challenging but crucial roles of the data team manager.

Data and AI projects often emerge from a business need or request, and translating this into concrete data-related problems is key to creating value for the organizations. This requires a fine understanding of the business and having in-depth conversations in order to better understand the different stakeholders’ needs and expectations. Being able to clearly define and communicate the business objectives and success metrics of data projects is critical to their success. 

2. Build Great Data Teams

Hiring and retaining great data talent is one of the biggest challenges to becoming a data-driven organization. When building a data team, don’t think that it’s just a matter of hiring the best data scientists that money can buy. One of the most important things to realize as a data team manager is that the data science process itself doesn’t solely revolve around data scientists — you need  individuals with different skillsets to support each step of the AI lifecycle. 

It’s also a matter of upskilling — leveraging existing staff from across the company, who can add their existing knowledge of processes and people to guide the team and make sure they are connected with all the lines of business. 

two women looking at a laptop screen

3. Evangelize a Data Culture Across the Organization

In addition to this overall balance between creativity and risk taking for data teams, it’s important for the data team to evangelize a data culture throughout the entire company. Creating a data culture means sharing and encouraging the following ideas widely:

  • Being data-driven and making data-driven decisions is a shared responsibility among everyone at the company (it’s not just the responsibility of the data team). This can manifest itself eventually through the development of a self-service data team.
  • Automation, predictive analytics, machine learning, and AI are a positive and important evolution for the company, not a way to eliminate jobs or responsibilities from other employees.
4. Build Trust Through Careful Governance Processes

Today, democratization of data science across the enterprise and tools that put data into the hands of the many and not just the elite few means that companies are using more data in more ways than ever before, and that’s really valuable (not to mention exciting for data teams who are seeing their role become more and more central to the organization). 

But it also presents new challenges - namely that businesses’ IT organizations are not able to handle the demands of data democratization, which has created a sort of power struggle between the two sides that slows down overall progression to Enterprise AI. 

Governance, when properly implemented, can improve trust in data at all levels of an organization, allowing employees to be more confident in decisions they are making with company data. Thus, understanding the need for a broader organizational shift towards a holistic approach to data governance, and being at the forefront of this shift, can allow data team managers to create trust in their teams’ efforts and unlock the path to Enterprise AI at scale that is responsible and sustainable.

You May Also Like

Why You Should Be Using Apache Spark + Kubernetes To Process Data

Read More

Is Yann LeCun the New Marie Curie?

Read More

How to Retain Data Scientists

Read More