|
How does a nonprofit fund-raising organization go about testing and refining its methods? An outstanding example of this type of database marketing is that of the Eastern Paralyzed Veterans Association (EPVA), which has used a database to guide its mailing very successfully for many years. Its database is managed by CSC Advanced Database Solutions.
To illustrate EPVA’s methods, the typical EPVA Christmas card acquisition was aimed at about 2.4 million people selected from 72 different lists. Once donors had given the first time, they were sent an appeal for a second purchase. This mailing was aimed at about 3.8 million first-time donors. EPVA broke its database down into five groups:
ND | New donors, who made their first gift after the year-end calculation |
AD | Active donors, who have made one or more gifts in the previous 12 months |
L1 | Lapsed donors who were not active in the last 12 months prior to year-end |
L2 | Lapsed donors who were not active in the last 24 months prior to year-end |
L3 | Lapsed donors who were not active in the last 36 months prior to year-end |
The EVPA also identified star donors, donors who at one time in their history had given to three out of four consecutive card programs and who had responded at least twice in the last 12-month period. This is a subset of the active donor category.
The EPVA’s goals for the future were as follows:
Determine donor profiles. Who are our donors? What are the attributes of a star donor? What are the differences between multiple- and single-gift donors? What characteristic should we be looking for when we rent prospect names?
Segmentation. What donors should be selected for what mailings? Establish giving patterns and program interrelation. What modeling approach should we pursue? Basically, we need to optimize our mailing plans.
Lifetime value. Which acquisition lists pull the best over the long haul? What effect do different test formats have on subsequent mailings? Would donors produce more revenue over the long term if they were mailed to less or more frequently?
Product analysis. Which products are the most effective in retaining donors? How do reminder mailings affect donor behavior? Are there groups of donors who consistently give to only one or two products a year, and what would happen to overall revenue if we mailed only those products to them?
Upgrading donors. What are the best methods of increasing donors’ gifts?
To achieve these goals, EPVA worked with CSC to undertake several projects.
Donor profiles. Create a sample of 400,000 records, with 100,000 in each of the following categories:
Star donors
Nonstar donors—single donors
Nonstar donors—multiple donors
Nonresponding rental files
Use the demographic data available on these records to determine the profile of these people, measuring their
Gender
Household income
Age
Housing type (own or rent and single family or multifamily)
Ethnicity
Presence of children
Years at present address
State
Urban vs. suburban vs. rural
DMA areas
Occupation of head of household
Education
Marital status
Number of adults in household
Create segments. EPVA planned to append Prizm codes to four groups of 10,000 each:
Star donors
Nonstar donors—single donors
Nonstar donors—multiple donors
Nonresponding rental files
These data would help EPVA to determine which lifestyle groups were more responsive to the EVPA message than other groups. The results would be used for future list selection and donor mailings.
Lifetime value. EPVA decided to use the data in its database to determine
Which lists pull the best over the long haul. To do this, it measured the response rate of each list used over the previous 3 years, comparing the results to determine whether the contributions were going up, going down, or staying the same.
The effect of frequency of mailing on the level of donation. To do this, it set aside a test group of 20,000 two-time donors and sent these donors only two mailings in the subsequent year. An exactly similar group of donors received four mailings. At the end of 1 year, EPVA could determine which group had the highest contribution.
Product analysis. EPVA decided to compare the performance of 50,000 star donors and 50,000 nonstar active donors to determine
What was the product that brought them to EPVA in the first place?
What products produced the best subsequent responses?
How did each group perform by product?
|