Sunday, March 1, 2009

Yes, Fundraising is About Relationships…No, Not Those Relationships!!!

For years, Development Professionals have repeated the same, tired mantra "fundraising is about relationships."

This mantra suggests that donors give only because they have a good relationship with the nonprofit's Development staff, and not because they have a passion for the cause.

There are at least three fundamental flaws with this so-called "truism."

First, if fundraising is about relationships, how come the average donor exists long before the average Development Director and remains active long after the average Development Director?

Second, characterizing giving as about relationships discounts donor intentions. If fundraising were about relationships, how come so many donors make "restricted gifts?"

In fact, a restricted gift for a cancer treatment ward for youth reflects a deep passion for care and treatment of cancer, rather than for finding a cure. Both are important, but the donor's value of treatment shines in such a restricted gift.

Third, if fundraising is about relationships, how come the business model for a nonprofit development office isn't havinge an army of Development Officers who meet with as many random people as possible to develop "relationships?"

For more information on donor motivation, we recommend an article by Paul Schervish, Ph.D (Boston College Center on Wealth & Philanthropy):

Inclindation, Obligation and Association: What We Know and What We Need to Learn about Donor Motivation.

When Development Professionals say "fundraising is about relationships," they could very well be right; the only problem is that some Development Professionals don't know which relationships are about fundraising.

At Arrowhead Management, we believe that the strongest relationships your organization can form are ones that can be identified by observing donors, their experiences, a host of data and several other characteristics of the donors' lives.

We call these experiences, data and characteristics "variables."

There are two types of variables: dependent variables and independent variables.

Dependent Variables
are the changes in behavior that occur as a result of independent variables. In a fundraising case, a dependent variable can be: that people give to your organization, how much money people give or how often people give.

Independent Variables
are characteristics that we can observe that MIGHT have an effect on the dependent variable. Independent variables can be: donor age, income, educational status, how long they have lived in the community or whether they are married.

Let's say you want to find out whether a donor's education level (Independent Variable) influences his/her decision to make several gifts each year to your nonprofit (Dependent Variable).

We combine these variables in the following formula, which by the way, we all learned in 8th grade Algebra 1 (but don't worry, no quiz at the end):


In this model:
Y=dependent variable
M=slope (or the relationship; we'll discuss in a moment)
X=independent variable
B=starting point on the graph

If you're looking at more than one independent variable, then the equation looks like:

Each independent variable is labeled "x1, x2."

Our next question is to determine whether the independent variable is "statistically significant" or not. One thing to know is that either an independent variable is statistically significant or it is not; there's no "in between" or "sortas."

Statistical significance is a way of asking "does this independent variable have an effect on the dependent variable or not."

You can determine whether a variable is statistically significant using a "fancy" formula on MS Excel (we'll discuss at a later time).

Now, it's important to understand at the outset that most independent variables are NOT statistically significant.

The next thing you need to know is that the relationship between the independent variable and the dependent variable is measured by an "R Squared (R2)" value. Independent variables that are statistically significant—or matter—have an R2 value that ranges between 0 and 1.

Click here for a Wikipedia definition of R2, but, briefly, the R2 is found from an analysis between the independent variables and the dependent variable primarily through regression analysis (fancy pants math that Excel can do for you).

An R2 that equals 0 means that the relationship between say, a donor's education level, isn't significant in determining how frequently s/he gives.

An R2 that equals 1 exists only in theoretical, math models (there are no perfect statistical relationships).

An R2 that is not between 0-1 indicates that the independent variable(s) is/are not statistically significant.

However, with statistically significant independent variables, the closer to 1 the R2 is, the stronger the relationship.

So, if your analysis reveals that a male donor who is retired, married, has lived in the community for 10 years and has attended three events your organization sponsored with an R2 of .8, that means 80% of his giving can be attributed to those relationships.

Imagine if you know this example to be the case and that you can go into your database and find other donors with the same attributes!

Now, in that case, fundraising is a lot about relationships…

No comments: