Case study: Federal government
Forecasting future work demand using advanced data analytics
Table of contents
A large federal health agency faced a critical challenge when confronted with increasing demand and not enough resources – impacting its ability to meet its mission. The agency wanted to use a data-driven approach to align its workforce with workload demand and equip themselves with more insights to advocate for additional funding. They partnered with Eagle Hill to develop their first-ever workforce capacity planning solution, utilizing predictive models and advanced data analytics, to accurately anticipate workload and manage resources proactively.
Goal
Make data-driven resource planning decisions using advanced data analytics techniques to forecast future workload and associated staffing needs.
Unconventional consulting—and breakthrough results with advanced data analytics
accuracy in forecasting future work demand
of core organizational functions covered by the forecasting models and outputs
data fields identified and used to increase the robustness of the forecasting model
The challenge
The organization was at an inflection point. It anticipated not having enough personnel to process workload, risking the agency’s ability to deliver on its mission. Without reliable data, and a way to reliably anticipate future workloads, leaders couldn’t optimize resources, make informed hiring decisions, or justify funding requests.
The roadmap to success using advanced data analytics
Eagle Hill began by working with our client to understand data reporting structures and identify information gaps. We reviewed existing data including workload, financial and time management records. Our team met with key stakeholders and hosted focus groups to understand how data was being used and to validate our understanding of the current business processes.
Based on our experience providing data analytics services, we recommended the following:
Implement an advanced data analytics forecasting model using predictive algorithms to anticipate workload for the next five years and justify future workforce needs.
Develop tailored performance dashboards that visualize data insights for faster and more informed workforce planning decisions based on immediate demand.
Utilize a monitoring dashboard to identify and correct underlying data quality issues in real time, making the process more efficient and improving confidence in forecasts and performance dashboards.
The path forward
With full buy-in from our client on our proposed solutions, we shifted our focus to implementation by:
Preparing the data & building a forecasting model to predict future workload demand.
- After validating data sources, we processed, cleaned and analyzed regulatory submission and time reporting data. To improve forecasts, we incorporated external economic and funding data such as GDP and inflation.
- With the clean data, we used Python and machine learning algorithms to develop a forecasting model that would prove to be 90% accurate in predicting future workforce demand.
Developing a performance dashboard and technology adoption plan, allowing key stakeholders to effectively use data insights.
- Utilizing Tableau, we collected, tracked, analyzed and reported on key performance data. Our ability to ask discerning questions about the data brought a fresh perspective on how to analyze key metrics. For the first time, agency leaders had business intelligence insights at their fingertips to inform workload and strategic planning.
- The success of the new workforce capacity planning tools hinged on bringing people through the transition smoothly. In this spirit, our team created a robust technology adoption plan to educate and build buy-in around the forecasting model and performance dashboards.
Building the foundation for flexibility and long-term success.
- When there was insufficient data available to account for specific process steps, we developed workarounds, including conducting assessments and manual tabulation through interviews with department heads.
- Eagle Hill built the model such that users can increase its accuracy with the accrual of additional data. As time passes and more data is collected, stakeholders can run validation checks on the model’s performance and use the information from those checks to validate model performance and determine the need for reevaluation.
- To get even more value from the forecasting model over time, our team provided recommendations that included standardizing data management and time reporting processes, refreshing training and updating IT systems, and data infrastructure and processes improvements to close identified gaps in required data.
Today, the agency is working more efficiently and effectively. Now they can plan for what’s ahead based on an accurate picture of future work and optimize resources to meet their mission.
