boodle Terms

Also called an Affinity Score or Fit Score, this is how similar a prospect is to other prospects who have previously shown an affinity for a nonprofit, campaign, cause, company, or product/service. This part’s important: this isn’t a likelihood or probability. You’re not looking for high numbers, you’re looking for similarity among numbers. An affinity…

Also called predictions, these are the entities the system recommends.

What we call a single row of data. For example, if you have a list of people who have donated to a nonprofit before all lined up in a spreadsheet, one row = one record. What exactly the record contains depends on how many features it has. It could just be a name, if that’s…

The process of ranking recommendations based on their score.

Also called a predictive multiplier, this is a way to measure how many more high-fit prospects are predicted to be in a set of recommendations.

Also called a predictive value, this is a way to measure how many more high-fit prospects are predicted to be in a set of recommendations.

A machine learning model (aka a fancy bit of code that’s trained to recognize certain types of patterns) that we use for predictive donor scoring — that process we do where we tell you who’s most likely to convert. Also called a guidon.

Also called recommendations, these are the entities a system recommends. In our case, it’s going to be the people who appear likely to give.

What we like to call a target data set, which is a stats phrase for a set of data you want to know more about. In this case, it’s gonna be prospects and potential prospects.

The process of connecting unique data like a person’s name, email, phone, and mailing address to create a single persistent identity that resolves to a record (aka one line of data) in a third-party database. Think of it like fleshing out the profile of a data point.