Lapsed or Dormant? Leveraging Statistical Analysis in Fundraising

[Editor's note: This article is based on the session "Lapsed or Dormant? The Impact of Fundraising Efforts on Donation Activity" held Friday at the 2011 Washington Nonprofit Conference.]

Bob hasn’t made a donation to a particular nonprofit organization in three years, while Mary last contributed to the organization five years ago. On the surface, it might seem that Bob is a better prospect for future donations, since he has contributed more recently. On the other hand, perhaps the organization could spark Mary’s interest again and get her to re-engage after her five-year absence. So, which of the organization’s donors should be the focus of its development efforts?

Addressing this issue highlights a critical yet subtle difference between a dormant donor and a lapsed donor. While a nonprofit organization can re-engage a dormant donor through its development activities, such as direct marketing, a lapsed donor does not respond to these efforts. Given that these fundraising efforts incur a cost, resources are better allocated to those donors who are dormant rather than spent on donors who have truly lapsed. The key is to distinguish between these two types of donors based on the information available to the organization.

Consider the case of a university engaging in annual fundraising drives. After a student graduates, the development office maintains a record of all of his donation activity. In each year since graduation, the university knows whether or not he has made a donation to his alma mater. The result is a sequence of observations for each individual, as illustrated below with a “Y” indicating a donation in a given year since graduation and an “N” indicating that the individual did not make a donation in a given year:

Year: 1 2 3 4 5 6 7 8 9 10
Graduate 1: Y N N N N N Y N N N
Graduate 2: N N N Y Y N N N N N
Graduate 3: Y Y Y Y Y N N N N N

All three individuals are 10 years from their graduation dates but differ in the frequency and recency of their contributions. Whereas graduates 1 and 2 have both made two contributions to the annual fundraising drives and differ in regard to when their last donations occurred, graduates 2 and 3 both made their last contributions five years ago but differ in how frequently they had made donations.

Based on these donation histories, who is most (and least) likely to contribute in the future? This question can be reframed into a related question: Which of these donors is most likely to be dormant, and which has probably lapsed and will never donate again? Simple summaries (such as recency and frequency) provide some relevant inputs, but they fall far short of offering any specific recommendations. Upon closer inspection, graduate 3 is the most likely to have lapsed, since he donated consistently early on but has gone completely cold since. It is much more difficult to compare the seemingly sporadic pattern of graduate 1 with the “off and on” pattern of graduate 2.

Rather than trying to flag lapsed and dormant donors using casual heuristics, formal statistical analysis can be brought to bear on this question. A recently published paper by Peter S. Fader, Bruce G.S. Hardie and Jen Shang, “Customer-Base Analysis in a Discrete-Time Noncontractual Setting,” examined exactly these kinds of trade-offs using annual donation data from a major public radio station. It showed remarkable accuracy in predicting donation patterns over a five-year forecasting horizon. While the model has some complex math within it, the ultimate implementation is a relatively simple Excel spreadsheet, requiring nothing more than recency and frequency as inputs. Many of the results and implications within the paper run quite contrary to the usual intuition that many nonprofit organizations have about the behavioral propensities (and thus future contributions) of their past donors. For instance, the paper found that recency is far more important than frequency in determining estimates of future donation patterns.

But while the paper offers accurate forecasts and useful insights about how to understand the underlying “drivers” of donation propensities, it stops short of offering any specific advice about which donors to target and when/how to do so. This is where a new paper by David A. Schweidel and George Knox, "Incorporating Strategic Direct Marketing Activity into 'Buy 'Til You Die' Models," comes in. Building on the same basic modeling platform used by Fader et al, this paper goes further to explore the impact of fundraising efforts on short-term and long-term donation activity. It explicitly captures the “dual causality” that often exists between donations and solicitation efforts — in other words, solicitation efforts clearly drive donation behavior, but past donations also impact solicitation strategies.

Once again, informal heuristics are unable to sort out these intertwined patterns, but a carefully crafted statistical model is able to do so. Schweidel and Knox show how an organization can gain specific understanding of how soliciting donors can not only influence the decision to make a contribution in response to a particular piece of direct mail, but also how it impacts their tendency to lapse. On one hand, fundraising efforts may deepen a donor’s attachment to an organization, but such activities may also irritate donors and trigger a premature end to the relationship. Striking the right balance here is obviously critical to any nonprofit, but it has not been examined very carefully on a regular basis.

Taken together, these two papers offer a powerful “one-two punch” for data-driven nonprofits that are interested in moving beyond casual “rules of thumb” to address key questions about future donation patterns and the best ways to enhance them using standard solicitation strategies. These methods have proven useful in other industries that also focus on the collection and analysis of detailed customer-level records (e.g., pharmaceuticals, telecommunications, financial services), and the carryover to nonprofits is surprisingly smooth and robust.

David A. Schweidel, Ph.D., is assistant professor of marketing at the University of Wisconsin-Madison School of Business. Peter S. Fader is Frances and Pei-Yuan China professor and professor of marketing at the Wharton School of the University of Pennsylvania.