Plan international is an international NGO aiming to advance children’s rights and equality for girls. They strive for a just world, working together with children, young people and communities to tackle the root causes of the challenges facing girls and all children in vulnerable situations. They drive changes in practice and policy at grassroots, national and global levels using their reach, experience and knowledge. Plan International has been building powerful partnerships for children for over 75 years, and is active in over 70 countries.
For more information about Plan International, you can visit their website.
Data and AI driven solutions are advancing rapidly and bringing tangible benefits to many private enterprises and public institutions, and there is high potential for AI also for NGOs - from increasing future efficiency of operations to broadening the reach and impact of their initiatives. While some NGOs have already started exploring or adopting AI initiatives, for many nonprofit organisations, AI may still seem out of reach. Building AI solutions often requires upfront investment in innovative, potentially experimental solutions, a fact that may, among others, pose a barrier to the adoption of AI for NGOs. (more info here)
Together with Plan international Belgium, we looked into a way to leverage the potential of AI to advance their mission of creating a better future for children. As an NGO, Plan international Belgium is reliant on funding, coming mainly in the form of either donations or child sponsorship. This fundraising process has been identified as one of the areas that could benefit from a more data-driven approach. With the goal of spending as much of the funds raised as possible directly on initiatives and thereby on Plan International Belgium’s main purpose, a more efficient and effective fundraising process will ultimately benefit everyone involved.
By gaining insights into the donor journey, efficiency and effectiveness of fundraising could be enhanced. With insights gained from a data-driven approach, the ultimate goal is to enable fundraisers to take better and more personal care of existing donors based on their needs, while spending less resources and funds in the process.
With a topic defined that could lead to significant benefits for Plan International Belgium, and therefore ultimately for the children, we looked into tackling the aforementioned issue of investment. In a partnership between Plan International Belgium, ML6 and the student association of UGent Everest Analytics, we have found a way of working to solve this challenge - the students of Everest Analytics building the AI solution together with expert support from ML6.
In order to make fundraising more effective and efficient, we will use machine learning to gain insights into donations, figuring out for which reasons donors are likely to churn, as well as identifying donors that are likely to stop their donations in the future. Knowing for which reasons donors commonly stop donating, and whether a donor is at risk of being lost, can make it possible to adapt the processes to donor needs and enable a more personal and resource-efficient approach to fundraising.
There are some main data and modelling challenges involved in the project. One of them is data imbalance, since there is a larger set of donors who will keep donating than there are donors who might be at risk of churning. Another challenge is feature engineering: in order to define the right features for the model, it is important to collaborate closely with subject matter experts at Plan international Belgium and incorporate their domain knowledge into the solution.
Project still ongoing.
One of the core values of ML6 is knowledge sharing - we strive to share our skills and practical experience not only with our clients and the open source community, but also with the future generation of ML engineers and data scientists. By working together with Everest Analytics, a student organisation of UGent, we are able to do exactly that, while at the same time enabling an NGO to become more data driven and set the first steps into their journey with AI.