Donations can be better secured when you know exactly who’s donating and why they’re doing it. Using data to get to know top donors better and more efficiently target portions of the population most likely to donate. Fund raising data is usually imbalanced–for every 20,000 constituents, less than a thousand might give, sometimes half that.

There is a data science approach to this imbalance, highly skewed data.

A data science approach could reveal that users who signed up with a particular internet provider email address were 5 times more likely to donate than those with a Hotmail email address. Whatever the reason behind this, this kind of discovery helps a fund raiser predict who will be generous as a committed donor and take the necessary actions.

Data don't comes from just one source. Someone need to be able to bring disparate data sources from CRM Database, website, 3rd party portals and other external data sources to have a 360-degree view to do data analysis for fund raising.

Caveats in Machine Learning

Feature selection, the process of finding and selecting the most useful variables in a dataset, is a crucial step of the machine learning. The caveat is Feature importance does not translate to importance in real-life. Link

It’s easy to build a model or derive insights from data. The real work starts when it needs to be interpreted, deployed, shared, managed, and adopted by an organisation. Link