What Linda and Susan can teach you about segmentation

March 8, 2018      Kevin Schulman, Founder, DonorVoice and DVCanvass

If you are interested in learning more about segmentation, we’re doing a free webinar on April 4th called “Why your donor segmentation and lack of real donor journeys cost you donations”.  This is a one–post version of what we’ll spend an hour on in that webinar.  Now to Linda and Susan…


Linda and Susan are both white, female Americans born on June 7, 1946.  And they are both donors to your disease organization.  That’s where the similarity ends.

Linda has the disease your organization fights.  She used your resources to learn about the disease and beat it back.  She’s volunteered to help others with this disease and with healthy living education.  She’s been a walk captain for you.  She’s donated 70 times in over 20 years of giving, offline and online, with steadily increasing gift amounts.  She last gave $40 four months ago.

Susan had a distant uncle who passed from the disease you treat.  She gave once when he passed away three years ago, then forgot about you.  About five months ago, she gave $30 again.

Should these donors be in the same segment?  Certainly not.

Linda is more committed.  As someone who has had the disease, she deserves different copy treatment at worst.  At best, she’d get a different appeal, with her donation going to support the things that a sufferer wants to address (likely different from a non-sufferer).  And the offer might be different – monthly giving, planned giving, or other upgraded giving befitting someone of her commitment.

Susan is a coin flip to make another gift, at best.

Yet through the lens of traditional recency, frequency, and monetary value (RFM) analysis, they are both 4-6 month, $25-49.99, multi donors.  Through the lens of demographics, you can’t slide a piece of paper between them.  And if you are using these factors to determine your segmentation, you are treating Linda and Susan the same.  You shouldn’t.

As Mark Ritson put it in Marketing Week, “If your segment is populated by different people who want different things, it is not a segment. It’s a joke and so are your skills as a marketer.”

So how do you create segments with meaning?  You can tell Linda from Susan using more advanced modeling than RFM, certainly.  Our friends at DonorTrends can model your file and reduce volume to those donors, like Susan, who are less likely to give to your organization.

That’s great for in/out segmentation – who gets what communication.  It’s a darned sight more predictive than straight RFM.

But you also need to write those communications.  Since Linda and Susan have different experiences with you and want different things from your mission, they should get different communications.

Transactional modeling can tell you who.  But you need to know why.

Donor identity is why.  People don’t donate to you because of who you are.  They donate to you because of who they are and how you help them be the version of themselves they want to be.  That’s different for someone who suffered from a disease versus someone who knew someone.  That’s different for dog people versus cat people.  That’s different for those who identify with the people you serve versus those who identify with your volunteers.

When you recognize those differences and play that back to your donors, you have a far better chance of retaining and upgrading your donors.  And you make them happier – win-win.