Our Work

Don’t make sweeping generalizations

“Don’t make sweeping generalizations,” my mother used to say to me.  In life that is pretty good advice which I’ve passed on to my own children.

But in the field of direct mail fundraising – or any type of mass marketing for that matter – sweeping generalizations are desirable and even necessary.  I’ll explain.  

Direct mail fundraising is not one-to-one communication.  Although we try to use personalization and customized ask amounts and other techniques like handwritten addressing to simulate one-to-one communication – it’s not.

Effective direct mail fundraising targets clusters of individuals with a message and layout that is designed to appeal to their perceived common interests, perceptions, desires and fears.  And so we must make certain assumptions and play the percentages.

Of course these assumptions should be carefully thought out and tested and retested.  Recently a demographic study was done for one of our clients that provided statistical support for many of these assumptions and generalizations we’ve come to use.  And so with documented evidence in hand to
back them up I’ll pass on some of them we often employ to lower costs and raise more money:

1) The older you are the more likely you are to give (until you reach age 90).  According to the study the response rate increases in a perfect curve with each age segment until you reach age 90 and above.  Donors in the age range of 80-89 had a response rate 64% higher than those aged 30-39.

2) No significant difference in response rates among men and women but men give larger amounts.  A 16% higher average gift according to this study. For this particular client there is also a two-to-one advantage in terms of the number of men on the file in comparison with the number of women.  But be careful with this one. Many times (especially with older folks from earlier generations) married couples write checks in the husband’s name but it is the wife that is actually making the giving decision.

3) The lower your income the more likely you are to give.  In our study this also tracked in an almost perfect curve with each increasing income bracket having a lower response rate than the preceding one.  Donors with a household income of $15,000 or less had a 40% higher response rate than those making $125,000 or higher.  As one might expect the average gift size did increase among the higher income brackets somewhat mitigating the lower response rate to the extent that there was no statistically significant difference in the Gross Dollars Per Piece Mailed between these two extremes of the income brackets.

4) Lawyers and Doctors are generally very poor prospects for donations.  This is an old
axiom in the business and in our study these two professions ranked dead last in terms of response rate.  In this study those who listed “Religious” as their profession ranked highest followed by “Retired” in second place and “Homemaker” in third.

5) The less education you have the more likely you are to give.  Those who completed just high school had a 21% higher response rate than those who completed graduate school.  Of course education level tracks closely with income level and so it is hard to determine which factor is the driver here.  Of course when you are making assumptions about who to target the “why” is less important than the “who.”

One big caveat that needs to be disclosed here.  I’ve chosen these categories to highlight because I’ve found them to be mostly true across a wide variety of different organizations over the years I’ve worked in this industry.  However, there are differences between organizations and you should always employ generalizations carefully.  Some of this is common sense.  For example, If you work in the field of education you might well find that the more educated donors respond better.

This is why direct mail fundraisers make good black jack players.  You might occasionally get lucky playing a hunch but the smart money plays the percentages. Good luck to you! 

Katapult MarketingDon’t make sweeping generalizations