Developed a system for predicting the likelihood of donors to donate in their will, utilising donor survey responses and demographics to calculate a likelihood score. This predictive model provides valuable insights into donor behaviour, allowing the team to tailor their marketing activities effectively.

Data Collection and Processing

• Donor Survey Responses: Collect detailed information from donor surveys regarding their interests, giving history, and motivations for charitable contributions.

• Demographic Data: Gather demographic information such as age, income level, location, and occupation to understand the donor profile better.

Predictive Modelling

• Feature Engineering: Extract relevant features from the survey responses and demographic data to build input variables for the predictive model.

• Model Interpretation: Building a data structure that allows the team to describe the scoring model utilising the data.

• Likelihood Score Calculation: Once the model is provided, calculate the likelihood score for each donor based on their survey responses and demographics.

Model Evaluation and Deployment

• Validation: Validate the predictive model using cross-validation techniques and metrics like accuracy, precision, recall, and F1 score.

• Deployment: Implement the predictive model to generate likelihood scores for donors in real-time.

Benefits

• Personalised Engagement: Tailor fundraising strategies based on individual donor likelihood scores to maximise engagement and donations.

• Resource Optimisation: Allocate resources efficiently by focusing efforts on donors with a higher predicted willingness to donate in their will.

By leveraging donor survey responses and demographic data to develop a predictive model, the team can enhance their fundraising efforts and foster long-term donor relationships effectively.