Opinion: How to Create a Data Culture for your Supply Chain

Recent experience in seven countries has taught us how crucial it is to create a data culture for supply chain by focusing on people as well as technology. This new, people-centered approach takes the idea of a commercial sector control tower and applies it to public health.

Leading companies from Unilever to Procter & Gamble have revolutionized their logistics over the past 10 years with this supply chain control tower concept. The basic idea is to set business goals, select indicators to measure, use data to track progress, and course-correct as necessary.

This private sector concept is now gaining traction in public health supply chains, where our job is to ensure the right people, technology, and processes are aligned to deliver health products to families and care providers when and where they need them. Renamed as a “visibility and analytics network” (VAN), this strategy is still fairly new, and ministries of health often struggle with key challenges: poor visibility into the data, gaps in processes, capacity, and an organizational culture around using data.

At JSI, we have decades of experience in helping governments track delivery of vaccines, medicines, and other health products, and we have developed digital systems in countries including Ethiopia, Pakistan, Tanzania, and Zambia. Yet we have learned that is not enough to have accurate, timely, and appropriate data — creating a data culture requires investing in people and process change, as well as technology.

Instead of treating logistics as a side job for busy health workers, we created IMPACT teams. The teams are deliberately made up of people across functions and disciplines, trained to interpret key indicators, such as stockout rates, encouraged to follow a problem-solving process, and empowered to use their data to make operational and strategic decisions that continually improve the performance of their supply chain.

Research demonstrates that pairing digital systems that enhance data visibility with this team approach achieves improvements in supply chain performance, thanks to better communication and such simple, local actions as timely redistribution of medicines. After IMPACT teams in Malawi achieved up to a 14 percent decrease in stockouts, the government scaled up this approach to all districts nationally. Kenya, Myanmar, and Pakistan are also expanding their pilot projects.

We’ve learned some key lessons in implementing IMPACT teams across seven countries to ensure they are successful and sustainable beyond the initial project:

1. Make data uncomplicated and actionable. In designing web-based dashboards, it is easy to be tempted by fancy graphics and animations — but they are useless if people cannot interpret the data or use it to make decisions. Data visualizations must be created using an iterative, human-centered design approach that engages IMPACT teams to ensure their needs are met. For example, through good data visualization, the IMPACT team in Guinea was able to identify supply imbalances and move products from overstocked to understocked facilities in their first meeting.

2. Offer basic training in data interpretation. Many health staff have limited experience in data mining and interpretation, problem identification, and root-cause analysis. They require significant training and ongoing support to build these skills. In Pakistan, IMPACT teams struggled to identify indicators and interpret the different graph types. We conducted additional workshops to teach basic skills in interpreting data visualizations, including comparing categories in bar charts, reading trends in line charts, and understanding the slices in a pie chart.

3. Develop and strengthen leadership. We have observed that IMPACT teams flourish with strong, effective leaders; where leadership is weak, they do not. In every situation, natural leaders will emerge, but given the importance, we have folded leadership development into our IMPACT team design, cultivating leaders as part of the process, rather than leaving it to chance. In Kenya, inSupply has implemented a leadership initiative, which has empowered selected individuals to lead IMPACT teams and has connected them for peer-to-peer learning.

4. Create networks across teams and levels. Connecting otherwise siloed teams better coordinates and aligns activities and helps resolve challenges. When health staff at lower levels of the system either do not have the authority to make certain decisions, or lack insight into the broader supply chain context, ensuring their ability to elevate issues is key to resolving them and motivating the team. Even teams working at the same level share information and work together. In Indonesia, IMPACT teams have helped individuals responsible for different activities, such as forecasting or distribution, come up with solutions they can jointly implement.

5. Recognize and celebrate progress. IMPACT teams require important changes in behavior, such as making evidence-based decisions and undertaking actions that are transparent, reinforce accountability, and align around common objectives. Celebrating small wins and recognizing achievements—small and large—help team members stay motivated, create a sense of friendly competition, and encourage teams to persist in new behaviors even before the results catch up with the changes in processes.

Of course, a VAN is of little use if a public health system cannot keep up the momentum after the life of a funded project. That is why sustainability is at the core of the model.

From the start, IMPACT team meetings are aligned with existing structures to ensure that few additional resources are needed to keep the program running. Team members are also coached to follow a structured approach for the meetings—interpret data, prioritize problems, conduct root-cause analysis, and plan for action—enabling them to successfully facilitate other routine meetings.

Making sure that such skillsets become common practice is what cultivates a culture of data and continuous improvement that can be applied to other supply chains and programs.

*This post was originally published on October 20, 2017, on Devex.

Written by Carmit Keddem and Yasmin Chandani