Model Behavior

Put simply, modeling means finding like characteristics about people that, when combined, enable you to predict future behavior. There are several inexpensive ways to model your data yourself (and some more expensive ways to use external data sources to help you) that can yield more revenue at lower cost from your direct-response program.

Modeling is best used when you have a specific question that you’re trying to answer, such as: “Who are my best donors to approach for an upgrade?” “Which petition signers in my database have the best potential for contributing?”

If you do not have a specific question in mind, it still can be useful to see if there are characteristics influencing the outcome of a mailing that were not in your selection criteria. For example, if donors were selected for a mailing based on recency, frequency and gift size, you might want to look at the results by region, the acquisition package through which donors were acquired or by cumulative donor value over their lifetimes.

Before going into specific cases where modeling can benefit your program, it is important to remember that there are things modeling will not do:

1. Modeling does not mean you can ignore direct-response best practices. For example, if you are modeling to eliminate donors from a mailing, test a sample of people you would have included had the model not eliminated them.
2. Modeling does not make up for a weak case for support. When times are tough, your case for support must be stronger as you are competing with many groups for limited dollars.
3. Modeling cannot capture the emotional element that leads to a contribution. The model can tell you who is more likely to give, but you still need to make a connection between that donor and your cause to motivate him to give.

These specific cases help illustrate some good uses of modeling. Each case begins with a simple question that an organization was trying to answer.

Case No. 1: Can I save money on a premium mailing without harming results?
Many organizations send an annual premium mailing — calendars, cards or other front-end gifts — to donors. These mailings have become essential elements of many annual mail plans, raising hundreds of thousands of dollars for nonprofit organizations of every kind.

However, premium mailings are expensive. In an effort to cut costs, many organizations are looking for ways to reduce the quantity of these mailings without decreasing revenue. Looking beyond basic RFM segmentation helped one of our client organizations.

This midsize nonprofit organization had always mailed its calendar to donors who had given a gift of $35 or more in the past three years. While all segments generated net income, many had a low-cost recovery.

The organization reduced the number of calendars it mailed as a front-end premium. But it still wanted to mail all of its current donors. So the organization developed an inexpensive, case-based appeal as an alternative package. The inexpensive appeal offered donors the calendar as a back-end gift.

The nonpremium donors were selected based on information about past giving to calendars, past giving to other premiums (like notecards), the acquisition package to which the donor first responded and the donor’s recency on the file, as well as traditional RFM. Using the additional criteria reduced the premium mail quantity by half.

The results? The calendar and inexpensive appeal together raised 50 percent more than in previous years, when the calendar was sent to all active donors. And the costs were significantly lower!

Case No. 2: Can I use acquisition merge/purge to improve donor performance?
A traditional merge/purge is used to clean up and dedupe the list of rented prospects. However, if you think outside the box a bit, that merge/purge process can reveal valuable information about your existing donors (and even your suppression file). The best part? This is a process you’ve already paid for!

One of our clients learned through past research that its donors supported, on average, 26 other nonprofits. Whew! Its goal then was to segment the next appeal by how many other organizations its donors supported — to learn how this “other” giving impacted appeal results. It used its merge/purge to learn this information.

Current donors and the traditional “suppressions” were processed as priority lists in the merge so the organization could see how many donors supported its organization only, one other organization, and two or more organizations. This information can be found in the list interaction section of the merge/purge report.

In the next appeal mailing, the organization used this information on “other giving” to segment the donors. It found that 39 percent of its donors did not appear on any other list in the merge/purge. While the response rate from these donors was slightly lower, the average gift was 23 percent higher than the control — causing us to surmise that this organization was a primary cause for these donors and a strong prospect for an upgraded ask.

Donors who appeared on one or more lists (almost 50 percent of the total donor file) produced a lower average gift but higher response rate. These individuals were targeted for additional mailings since they proved to be very responsive to mail appeals.

The most surprising information concerned the organization’s “suppression” file. For many groups, these files become catchalls for a variety of special groups — mail only once a year, no premiums, hastily reported deceaseds, donors marked as “do not mail” by major-gift officers and a variety of other records get put in this category.

By segmenting out the true “do not mail” people who requested no mail, we separated them from other special-handling groups, who could potentially be good prospects.

In the end, 11,000 “suppression” records were giving to other organizations — including people who had been marked as deceased by this organization! These individuals were sent the acquisition, generating a high response and a net profit. The lesson learned was that there may be people you aren’t mailing but should be!

In the same way that a suppression file can be “modeled” through merge/purge, you can assess viability of other nondonor groups on your database. Event participants, volunteers, petition signers, etc., are not typically good direct-response prospects. But you can include these lists in the merge/purge process to find out how many are giving to other groups in your market.

If the percentage is similar to the other lists in the merge/purge, it could be a viable list. If it is significantly higher, you won the lottery. If it is significantly lower, you probably should only mail those who matched outside lists.

Case No. 3: If I have to cut acquisition quantities, how do I do the least harm?
It is absolutely critical to continue acquiring new donors even when the economy is bad. If you stop acquiring new donors, your file size will begin to decline, and your long-term revenue will as well. Unfortunately, some groups find themselves with no choice.

Outside modeling tools can help narrow your acquisition mailing. There are cooperative database tools available that you can use to filter the acquisition once the merge/purge is complete. For example, one of our clients uses it to eliminate prospects who are only giving gifts of $5 or less to other organizations. Even though the organization still pays for the name, it saves the cost of mailing an acquisition and saves the expense of trying to renew a $2 donor who likely will never upgrade.

You also can use outside tools to find the donors on your database in whom you should be investing the most. You can get information about age and basic wealth indicators relatively inexpensively (to help target planned-gift mailings), and other products can help you find donors who may contribute more. You also can have regression modeling and other processes performed on your data to help you segment better. However, these tools can be expensive and should always be tested. FS

Lynn Mehaffy is an account supervisor and Tiffany Neill is a partner at Lautman Maska Neill & Co.. Reach them at lmehaffy@lautmandc.com and tneill@lautmandc.com, respectively.