RFM analysis is a great first step to analyzing and grouping your donors. By summarizing their giving history and ranking by Recency, Frequency, and Monetary values, you can cluster donors into smaller comparable groups. This can help with identifying possible future prospects, searching for a target donor audience, and allow for more personal communications to each group.

All organizations are striving for continuous improvement, and getting feedback from your stakeholders is essential to that goal. One of the best ways to get that feedback is by surveying your constituents - students, families, employees, alumni and/or donors. Learn how to conduct your own survey research, even on a modest (or non-existent!) budget, and collect actionable data to help you improve your organization.

Learn how MIT is taking a strategic and tactical approach to data management in preparation for future technology transformation. Leading from critical needs of the organization, MIT is developing a data classification strategy to inform data migrations, reinforce data standards, and leverage established project management principles to strengthen its data management practices as they respond to the changing needs of their constituents and transform their business.

As personalized and curated experiences continue to take over digital and commercial spaces, donor engagement teams must continually innovate to keep up. All-in-one solutions can provide small teams with scalable platforms and comprehensive tools that drive results, but they often come with limitations in product development that don’t meet the needs of individual clients. Over the last year, Georgia State University’s Donor Engagement and Communications team has used data analysis, strategic engagement and personalized outreach to develop an internal donor experience program focused on long-term results: the acquisition and retention of mid-level donors with the aim of carrying them through the “murky middle” and establishing them as our next generation of major donors.

Much of the conversation about improving predictive models focuses on finding the optimal model technology, such as linear regression, logistics regression, classification trees, etc. We argue that there should be just as much focus on Exploratory Data Analysis (EDA) techniques by exploring, diagnosing and transforming the data - before, during and after the modeling process. We will also demonstrate how to apply different Exploratory Data Analysis (EDA) techniques to find optimal transformation and identify outliers.