Last month we heard from USC about their experience in creating a model for donor affinity and this month we have the pleasure of hearing from Kathy Mills, Senior Donor Identification Analyst, at Baptist Medical Center who shares their approach for a similar project at Wake Forest.  

In 2013, our prospect management/development team here at Wake Forest Baptist Medical Center embarked on a plan to create an affinity score for our non-managed prospects. The goal was to use this score as a tool to find lower-level donors who were shaping up to look like our major gift donors and who should be assigned to a major gift officer.

First, we brainstormed about characteristics of our major donors and criteria that would indicate someone had an affinity for our organization. An APRA Carolinas meeting where fellow member Patrick O’Toole presented on the topic further guided our methodology. After lots of conversation, we decided on the following:

•         Consistent giving
•         Gifts of stock
•         Gifts through a donor-advised fund or foundation
•         Planned gifts
•         Attended 10 or more events (lifetime)
•         Alum of the medical school
•         Alum of both Wake Forest University’s undergraduate and medical school
•         Current or former board member
•         Hospital volunteer

We knew that some of these characteristics should carry more weight than others, but we were not sure how to determine those weights, so we consulted with statisticians on our faculty. We sent them a list of 800 donors, including some of our very highest-level donors, and included their total household giving and which of the above characteristics they had.

The statisticians analyzed the file and made their score recommendations. After a little tweaking, we finalized the figures and determined the maximum score a prospect could obtain. We had our systems analysts overlay the scores on our sample file of 800 donors and analyzed the results. We were delighted to find that the donors we expected to have the top scores did indeed score very highly. The scoring system appeared to be working.

Next, we had our systems analysts overlay the scores on a file of about 29,000 unmanaged donors, with the data refreshing every night. As with our previous test, those with the highest scores also tended to have higher household giving levels and alerted us to those who were shaping up to look like major donors. What surprised us, however, was how few donors scored highly. We uncovered very few hidden gems that could be assigned immediately to major gift officers.

This exercise indicated to us that we needed to do a better job engaging our annual funds donors to help move them through the pipeline. We are currently looking at ways to do that, including digital communications, event invitations, better stewardship for annual giving donors, and other methods.

We also realized the need to have special gift officers on our team to help move those with moderate scores to a higher level in which they would be ready to be assigned to a major gift officer. We are now in the process of hiring these gift officers.

We took a snapshot of what our unmanaged donors’ scores look like today. With the enhancements we will be making to our program, including increased engagement and special gift officers, we are hoping to see the overall scores increase over the next few years.