Harnessing Data Science for MSF’s global workforce
Strategically managing a global workforce of over 45,000 people to deliver medical care in over 70 countries is a tremendous and unique challenge. This challenge is compounded by MSF’s rapid growth and the unprecedented complexity of the medical humanitarian landscape – not only do we need the right people now; we need an ever-increasing amount of experienced staff to address new and rapidly changing internal and external demands.
Tell us about the problem you are trying to solve.
When faced with human resource (HR) challenges, the conventional approach is to engage in debate and rely on experience and intuition. However, this is no longer enough, as evidenced by the number of gaps in key positions in the field, such as medical coordinators (MedCos). This problem is compounded by our inability to forecast the number of field people we need to recruit to meet future operational needs and the constantly moving nature of MSF’s work and staff.
A second problem is limited evidence on what drives the needs and behaviors of our global workforce. Metrics on attrition and retention in an employee’s lifecycle are largely unavailable. It also raises a question on inclusiveness – can the voice of all staff be included in our strategic planning if we have limited insight on what that voice is? This is especially true for national staff.
What is your solution? What motivates you to work on addressing this problem?
Leveraging the power of data science, we can develop tools and deliver insights to HR teams to make more informed, objective, people-centered decisions. Data science is a multidisciplinary field – it’s math, statistics, and computer science, applied to data to uncover new insights, measure the effect of policies and decisions, and challenge (or prove) longstanding beliefs and practices. It can lead to more effective decision-making.
Organizations that invest in People Analytics functions translate effective people decisions into gains in efficiency, productivity and efficacy. Some organizations have invested in data science and People Analytics and use it for everything people related – from understanding and proving what makes teams effective, to using Artificial Intelligence (AI) to power internal talent development. The added value is proven and measurable. We believe we can see the same impact and return on investment for MSF.
People are at the heart of everything MSF does, and people decisions impact every layer of the organization. Every improvement we make in HR decision-making improves our social mission.
What have you done so far and what results have you achieved?
The HR Pipeline Tool, our deliverable in Phase 1, was the first time MSF HR data was used to create a tool that will fundamentally changes the way we approach workforce planning to be evidence-based and proactive. The tool changes the conversation between Operations and HR around the allocation of first departure international staff vacancies in the field. It helps define how many international staff doctors and nurses need to be recruited, placed, and developed to have a strong pool of MedCos in future years. The tool can help build a strong pool of MedCos in future years, as we now have data on how long that development takes, as well as insights on attrition and retention at every step in an employee’s career journey. This provides HR leadership with the capability to have a broader conversation and use the Pipeline Tool to simulate future scenarios to fully articulate the outcome of unmet HR needs, such as gaps in the field.
We are also evaluating national staff data to better understand this critical segment of our workforce. Our goal is to gain a deeper understanding of national staff needs and data to think differently about our staffing models, expatriation policies, retention, attrition and professional development.
What challenges have you faced? What lessons have you learned?
In the early stages of the project, it was hard to define what a People Analytics program would look like in practice. Attending conferences, reading articles, and talking to industry experts provided interesting perspectives, but formulating a vision for how this would look like at MSF was a unique challenge. From that challenge we learned that there is no “one size fits all” solution, and that it will be a process of trial and error. Getting comfortable with this fact helped us be flexible, tempered our expectations and empowered us to be bold in our approach and unafraid to take risks.
We also learned that data science work does not have to be extravagant and expensive to be useful, and answers to most questions can be found in simple analyses using free open source tools. The Pipeline Tool is a great testament to that, because the methodology underpinning the tool is not particularly complicated. In fact, it’s quite the contrary – Our tool uses a simple data science solution to a very complicated problem. Our next challenge will be to keep these lessons learned at the forefront while we advance and grow and take on bigger challenges. We want to embrace simplicity while still taking our solutions to the next level. We want to remain agile and comfortable with trial and error as the project grows in scope and complexity.
What is the expected long-term impact of the project? How will this project improve MSF’s lifesaving work and the lives of MSF’s patients?
The long-term impact of the project will be that HR teams and leadership will have the tools to fully realize the maximum potential of MSF’s human capital and leverage it to deliver on our social mission. While the monetary value of avoiding gaps in key positions in the field is very difficult to quantify, we know that a single gap in a key position in the field presents increased institutional risk, efficiency loss, and adds a tremendous burden on already strained field teams. For example, despite our massive recruitment efforts and our goals, there is a perpetual shortage of coordinators in the field. If we are able to provide HR teams with not only accurate recruitment and placement metrics, but also provide them with knowledge on how long it takes to develop coordinators, we can avoid some of the efficiency loss we are seeing on the recruitment side and give them the knowledge they need to execute the best strategy.
Given that coordinators play a key role in the design, implementation and evaluation of medical, logistical and administrative project activities, team management, staff safety and security, it is imperative that gaps in these roles are avoided. If we can plan ahead and make timely, informed, strategic and proactive decisions about these roles, it can have a positive impact on MSF staff in the field and help us better meet the needs of our beneficiaries. For a MedCo, this could mean identifying new medical humanitarian activities and designing and implementing them, addressing hidden or unforeseen medical needs of the populations we serve, thereby saving more lives. This is where we believe our work will make the most impact. We are not only arming our HR teams with the tools and insights to prevent loss – we are providing them with tools that can help teams extract the maximum value from our global workforce.
What have staff said about the project?
The support and enthusiasm for this project has been unlike anything I have seen in my 5+ years at MSF. I think this stems from how great the needs, and how difficult some of the questions we are attempting to answer are. The response and support at all levels has been overwhelmingly positive, and MSF’s HR teams and leadership have demonstrated an extraordinary openness to change. We have received feedback that simply presenting what we have done with the Pipeline Tool plants important seeds and introduces a new language, raising the opportunity to think about old problems in a new way. Once people understand the insights that can be extracted from international staff departure history data, their curiosity piques and they become excited about answering longstanding HR questions. At the leadership level, we often hear how this work underscores the importance of people policies and impacts MSF at every level.
“The pipeline tool has a lot of insight already that we can use in our day-to-day work, and it has the great potential to develop further. Imagine if you were about to make your annual goals for recruitment, and you were able to say, “Okay, I need 50 of this profile, and 100 of this profile because that will ensure that we have enough coordinators in two, or four, or five years. I think that is truly what’s going to make the difference here. It’s also going to give us much better visibility into our workforce, and that’s going to drive informed decision making. Ultimately, it’s going to allow us to use the full potential of our entire workforce. I think it’s also going to allow us to provide answers to a lot of the questions that we’ve been asking ourselves for years and years. By using the knowledge and the data that we have, suddenly we’re able to back up what we’re trying to do. We’re able to make strategies based on solid data AND our own HR expertise.” – Alexander Buchmann, MSF-USA Field HR Director
Are there any interesting partners that you are collaborating with?
We have many interesting collaborators. Dr. John Boudreau, a TIC Selection Committee member and the project’s advisor, is the Research Director for USC’s Center for Effective Organizations and a Professor of Management and Organization at Marshall School of Business. He is recognized worldwide for his breakthrough research on the bridge between human capital, talent and sustainable competitive advantage. Having John’s experience so closely follow the project and provide such relevant feedback has been a game changer. We have also worked with the University of Pennsylvania Wharton School of Business’ People Analytics program. They helped us define the problem statement and conceptualize solutions. McKinsey’s People Analytics and Measurement team also supported us, helping with the analytics and providing coaching in understanding the building blocks work on a global scale are. We are also building collaborations with academia for support on qualitative research design to ensure the tool is valid and reliable. The intersectional cooperation and collaboration on this project has been incredible. This is largely driven by a natural curiosity of what the data is revealing from one OC to another, as well as partner sections reaching out for support with analyses that answer questions specific to their HR practices.