Apra_573924-21_PD2021_WB-941x261-SaveDate.jpg

Boost Annual Fund Effectiveness with Python's Data Science Ecosystem

Track: Data Science

Session Number: 4078
Date: Fri, Jul 30th, 2021
Time: 12:00 PM - 1:00 PM

Description:

Annual funds often are troubled by external issues—the need to activate lapsed donors, gift sizes not increasing alongside capacity, secular decline in participation rates—that threaten the health of the fundraising pipeline. Annual giving is instrumental to a development office since the appeals they work on raise current use dollars while keeping donors engaged with the organization. Consistent annual giving helps to establish giving patterns that will ultimately result in donors progressing through the pipeline to become major gift prospects. This presentation demonstrates the ways in which researchers can mitigate these challenges through data science, distilling actionable insights from our databases to help improve annual fund efforts.

Annual fund teams already use data on a daily basis to help formulate strategy so they are open to higher levels of analysis. This session will walk through how to clean, explore, and model annual giving data using Python and select open source libraries in pursuit of two main goals: one, identifying donors most likely to churn (i.e. not give next year); and two, identifying current donors most likely to move up a giving tier. With these groups of donors identified, annual fund teams can tailor messaging more precisely to reduce donor churn, increase participation, and boost revenue.

Attendees will have access to the code used in this project, as well as a curated list of resources to get started with the powerful Python data science ecosystem.
Session Type: Breakout Session (60 Minutes)

Learning Objective #1: Attendees will apply data cleaning and exploratory data analysis (EDA) techniques to annual giving data using Python.
Learning Objective #2: Attendees will build machine learning models to derive insights from annual giving data using Python.
Software or Vendor Tools: Yes
Software or Vendor Tool Details: Python, Pandas, Plotly/Dash, scikit-learn, PyCaret
Primary Competency: DS:Competency 3: Data Manipulation Skills
Secondary Competency: DS:Competency 5: Visualization/Reporting Techniques and Competencies
Tertiary Competency: DS:Competency 1: The Data Science Umbrella
Intended Audience Level: Level II
Recommended Prerequisites: familiarity with fundamental programming concepts, fluency with advanced Excel techniques (e.g. pivot tables) may be helpful, knowledge of integrated development environments (IDEs) or notebooks (e.g. JupyterLab) is beneficial
Shop Size: All Shop Sizes
Session Type: Breakout Session (60 Minutes)

Learning Objective #1: Attendees will apply data cleaning and exploratory data analysis (EDA) techniques to annual giving data using Python.
Learning Objective #2: Attendees will build machine learning models to derive insights from annual giving data using Python.
Software or Vendor Tools: Yes
Software or Vendor Tool Details: Python, Pandas, Plotly/Dash, scikit-learn, PyCaret
Primary Competency: DS:Competency 3: Data Manipulation Skills
Secondary Competency: DS:Competency 5: Visualization/Reporting Techniques and Competencies
Tertiary Competency: DS:Competency 1: The Data Science Umbrella
Intended Audience Level: Level II
Recommended Prerequisites: familiarity with fundamental programming concepts, fluency with advanced Excel techniques (e.g. pivot tables) may be helpful, knowledge of integrated development environments (IDEs) or notebooks (e.g. JupyterLab) is beneficial
Shop Size: All Shop Sizes