By: Rebecca Herrington
Data-driven decision making may sound like jargon, but when approached honestly, data-driven decision making is a tangible practice to support more effective and efficient planning and operations. While the phrase may be overused and often lacking in the rigor of its application, this blog will hopefully demystify and provide practical guidance on data-driven decision making. This is especially important in light of the U.S.’ Foundations for Evidence-based Policymaking Act of 2018, the Assessment, Monitoring and Evaluation Policy for the Security Cooperation Enterprise, and continued efforts to improve the uptake of evidence in the international development sphere and beyond.
Expanding Application Based on Proven Value
Data-driven decision making is not new. The private sector has been using this approach through utilization of quarterly reporting, regular forecasting, and various business intelligence tools. This approach is found in Mobile Network Operators working to understand the cash flows and activities of agents supporting their mobile money services, in market assessments that help firms determine where best to make their investments, and in application of user testing findings to adapt innovations. But data-driven decision making is not only relegated to the private sector. Part of the utilization for businesses is in determining customer needs, trends, and how to best meet those needs to serve the customer. But what if the customer is a single mother with three children in Nigeria? Is data-driven decision making equally applicable when the ‘customers’ are villagers facing intercommunal conflict or mobile phone users trying to use a digital identity platform to access their nearby health clinic?
YES! The utility of data-driven decision making in international development to improve aid effectiveness has emerged over the past few years through uptake of the Collaborating, Learning, and Adapting Framework and its tools and approaches. It has been emphasized through the need to track progress and contribution towards countries’ journeys to self-reliance. And as mentioned, it has been increasingly mandated by the U.S. Congress across agencies, sectors, and contexts.
Data-driven decision making is accessible to everyone, but does require a time investment and a willingness to work through a series of actions to institute the benefits of data-driven thinking into day-to-day management. The core steps for data-driven decision making are captured in the process graphic on the left and described in the steps below. Remember, it is essential to approach data-driven decision making as a whole. Simple articulation of vision, or analysis without application of findings will not provide results and can be a waste of earlier efforts.
Step 1: Develop Your Strategy
Strategy development is the process of setting a mission, vision, priorities, and associated action planning that enables organizations (or specific activities) to articulate their purpose and the most effective pathway to act on that purpose. Without a strategy, there is nothing to guide your data-driven decision making efforts and ensure that data collection and use is streamlined and efficient. Click here to learn more about strategy development and how Headlight can help you build yours.
Step 2: Determine What You Need To Know
After you’ve developed your strategy, it’s time to design any monitoring and data collection you might need. What information would best inform you of how you’re performing against your objectives? What information would let you know how the context is shifting and how those shifts are impacting your work? Is there other information that would provide a more nuanced picture of the implementation quality of your work or capture early lessons learned? Figure out what metrics are feasible to capture on a regular basis and will provide you the most actionable data. Make sure you build the monitoring systems necessary to collect and process the data up front, as well as set aside resources for continued data collection efforts.
Step 3: Collect, Clean, And Analyze Data
Once you have your monitoring system set up, put it in action! Collect data on a regular basis, whatever frequency makes the most sense for the decision making schedule of your organization. If you conduct business planning in October for the next year, make sure you have all your data collected, cleaned, and analyzed by September. Most organizations collect data on an ongoing basis and clean and analyze the data on a quarterly or semi-annual basis. It is important to note that if you are working on an innovation or going through a growth period, you may need to increase the frequency of this step, compared to more typical operations. Make sure you have someone who is able to clean the data once it is collected to make sure it is captured accurately and is appropriately recorded to facilitate analysis. And lastly, conduct analysis. Look for trends on what is working and what is not working and why.
Step 4: Make Adaptations Based On Evidence
Finally, put the data in action. Data-driven decision making means basing decisions regarding resourcing, implementation, priorities, and so much more on what the evidence says. What is effective? What efforts are demonstrating improved return on investment? What enabling factors are catalyzing your work? Be willing to admit failure and depend on ongoing data collection to provide additional evidence on the effectiveness of any adaptations or pivots that you implement. Look at ways to further leverage what is working to replicate in different contexts or scale your work. And remember, data-driven decision making is cyclical and best when institutionalized. Get the whole team on board and familiar with depending on data to make decisions, pitch ideas, and be agile in the complex environments in which we work.
Data-Driven Decision Making Support
Headlight would love to talk to you about supporting improved data-driven decision making in your work. We have the breadth and depth of expertise, experience, and toolbox to tailor-meet your needs. For all questions, inquiries, and sharing of requests for proposals please email <email@example.com>.