Case studies

At the heart of Centraide's activities lies an important reality: its employees put in an enormous amount of time and effort to find donors. The organization seeks support not only from individuals but also from companies, which can often contribute on a much larger scale. Finding new corporate donors plays a crucial role in ensuring the sustainability of revenues and achieving Centraide's high aspirations for its community. 

However, despite the dedicated efforts of its employees and volunteers, these outreach initiatives sometimes result in refusals. These situations, besides being disappointing, consume valuable time that could have been spent approaching other potential companies who might have provided financial support to Centraide. This is precisely the challenge Vooban set out to address: optimizing the search for new potential corporate donors to maximize Centraide’s chances of securing new contributions.

The project we developed centers around a key objective: identifying the companies that are most likely to become donors. We used several interrelated factors to achieve this, including lists of companies, their sector of activity, location, size, history of donations to Centraide, and so on. The goal of the project was to optimize the time and resources of Centraide employees so that they could focus their efforts where they will generate the most value for the organization. 

Using classic machine learning techniques, we developed a tree-like artificial intelligence model. We call it this a tree because it represents decision-making in the form of a tree structure! Each branch corresponds to a decision based on specific characteristics or variables, and each leaf corresponds to a conclusion or prediction.

The advantage of this approach is that it makes it easy to understand how decisions are made. In this case, it's particularly useful for explaining why a company is more likely to donate and what characteristics make the difference.

By gaining a better understanding of the factors that make one company more likely to donate and applying these insights to a database, we've been able to identify new, high-potential companies that were not yet on Centraide's radar. This will enable Centraide to focus their efforts and prioritize these promising new prospects.