Plan Canada acquires new monthly donors through a variety of marketing channels and external vendors. Each acquisition source comes with different costs — and the donors acquired through each channel have different behaviors, including:

• Varying monthly donation amounts
• Different long-term churn rates
• Different patterns of ongoing engagement

Comparing the ROI of acquisition channels was complicated by a fundamental challenge: You can't directly measure the lifetime giving value of a donor until they churn — which could be years or even decades later.

Without a way to predict donor lifetime value (LTV) early, Plan risked under- or over-investing in acquisition channels, leading to inefficient use of fundraising resources.

CVM partnered with Plan Canada to develop an AI-powered model that estimates the predicted lifetime value of donors based on early signals.

Key aspects of the solution included:

• Predictive Modeling: Built a machine learning model to estimate donor LTV using factors such as acquisition channel, initial giving behavior, demographic information, and engagement patterns.
• Channel ROI Analysis: Enabled Plan to calculate the estimated return on investment for each acquisition vendor or channel shortly after acquisition — without needing to wait years for actual churn data.
• Data Integration: Combined internal donor records with vendor-supplied acquisition data to create a consolidated analysis dataset.
• Stakeholder Collaboration: Worked closely with Plan’s fundraising teams to align on model methodology, incorporate domain expertise, and ensure full buy-in before results were shared.

The model was built using Python and integrated with Plan’s existing donor management infrastructure, allowing for ongoing updates and refinements.

• Smarter Acquisition Decisions: Plan gained a data-driven basis for evaluating and optimizing its acquisition spend across channels.
• Improved ROI: Resources were reallocated toward higher-performing vendors and channels, improving fundraising efficiency.
• Faster Insight Generation: Estimated LTV could be assessed within months of acquisition rather than waiting for actual churn to occur.
• Stronger Stakeholder Confidence: Transparent methodology and close collaboration helped overcome initial skepticism and fostered greater trust in analytics-driven decision-making.

• AI Can Unlock Strategic Value: Predictive modeling enables better decisions even when full outcome data isn't yet available.
• Stakeholder Buy-In Matters: Engaging fundraisers and marketers early ensures adoption and maximizes impact.
• Continuous Refinement is Key: As more real-world data became available, CVM worked with Plan to retrain and improve the LTV model over time.