Chances are good if you are in the fundraising field that you have heard the term “fundraising analytics.” You’ve probably also heard the terms “data mining,” “donor modeling,” “reporting” and “prospect identification,” too. Do these terms mean the same thing? What are the differences among them?
I asked Marianne Pelletier, who leads the HBG Analytics team, to help me put together a series of short articles designed to make sense of all of this. In each, we will describe the method and give examples of how they can be used. To begin our series, we discuss data mining.
Let’s begin with a case study:
The fundraising team at a university is having a problem with donor retention. Every year the university must acquire a significant number of new donors to offset the nearly 50% of donors they lose from the previous year. They need to find an answer to the question “why are we losing so many donors, and what can we do to keep them from leaving?”
They decide to use a technique called data mining to find out.
What is data mining?
Sometimes we don’t know what we don’t know. In those situations, it may be best to explore a little to see if we can find answers (and perhaps even see the questions we need to ask). Think of it as discovering what’s in a new mall by walking the length of it: data mining is like shopping in the data mall. What will we find when we look in each store?
Data mining sifts data back and forth until it finds natural breaking points, puts together associated characteristics and then lays out what it finds. For example:
“The donors who give the highest gift amounts are married male alumni. They’re in their 50s. They live along the east or west coast. They have a job and children that we know about.”
Data mining delves deeper to find relationships between many characteristics
The good news is that data mining also names the characteristics for the second best segment, and the third best, and so on, so a university could use it to find out various solutions to their attrition problem. The characteristics they see might look like these:
“The donors who stop donating tend to leave between their 5th and 6th years of consistently donating to the annual fund” So we can work harder to retain them before year 5!
“The donors who stop donating tend to be married alumni males in their thirties” What special incentives can we offer to that group since we know that married men in their 50s tend to be our largest donors later on?
Data mining can also be used for:
- Determining the best solicitation methods for donor acquisition, renewal, or upgrade
- Measuring the characteristics of event attendees who later become donors
- Understanding the clusters of members/grateful patients/families/alumni/docents who respond better to e-mail, direct mail, phone calls, or social media
- Finding the best pattern for the cultivation/giving ladder
- Adding or dropping solicitation methods, or venues
- Assessing timing, including how long it takes to successfully solicit gifts at different levels
What else can data mining do?
Have you ever walked into a store that has section after section of fun things for purchase? A department each of clever t-shirts, gifts that your best friend would love, beautiful hand-crafts, and more – things that are so perfect that you’ve picked up an armful of things and you need to find a basket to dump them all in.
Data mining is like that. It can also be for:
- clustering like-minded, or like-attributed prospects for a cultivation dinner
- breaking long-held “truths” about your donor base, such as “our best athletics prospects are male football alumni.” What if that’s not actually true?
- looking at what statistics calls “interactions” – the combination of characteristics that make good prospects, members, volunteers, trustees, etc. For example: Married prospects and/or prospects living in rural areas show a mild relationship to loyal giving. However, prospects who are married AND live in rural areas show a strong relationship to loyal giving.
- determining which group responds best to email and which to social media.
What do you want to know?
Our series continues next Thursday, September 12 when we will discuss Donor Modeling.
Do you have questions about data mining or would you like to see how it can work for your organization? Email us for more information at info [at] helenbrowngroup [dot] com.