Predicting Sustainer Retention

March 25, 2019      Kevin Schulman, Founder, DonorVoice and DVCanvass

Monthly givers should be like an annuity or bond – an initial investment pays steady dividends over time.  In reality, sustainers are great, but they are not Ron Popeil’s “set it and forget it”: there’s far more investment of time and energy required to make sure a donor stays with you for the long-term.

What if you could know at the point of acquisition whether someone would retain or not?

Using data from DonorVoice’s sustainer retention platform, which gathers donor feedback and point of acquisition and immediately after, we can see what’s most predictive of retention.  Here are our contestants:

  • Commitment to the organization
  • Demographics (in this case, age)
  • Identity (in this case, whether someone has a direct connection to the disease)
  • Satisfaction with the experience

Coming in in fourth place is demographics as least predictive.  In this case, 38% of those 40+ retained versus 30% for those under the age of 40.

Since I’m on record saying demographics are the worst form of segmentation some readers may think I’m using this to bury demographics again; in this case, though, I aim to praise them.  A eight-percentage-point increase in retention is nothing to sneeze at.  More importantly, though, a demographic like age is one of the few you can spot at a distance.  Even your best face-to-face canvasser won’t be able to tell whether someone is committed to your organization from their gait.  And unless you are canvassing for Smurfism and see Violet Beauregarde walking down the street, you won’t be able to tell connection to the organization.

So demographics have their place, not as the be-all-end-all, but rather as a screening tool for canvassers.  If you were a canvassing firm focused solely on retention (which sadly is like saying “if you were a unicorn”), you would screen for older donors.  Most already do this to some extent.

In third place is identity. “What?”,  you say.  The Agitator isn’t putting identity first?  Well, we go where the data goes.  In this case, those with a direct connection to the disease had a retention rate of 46% and those without had a retention rate of 37%.  One caveat: this was before this nonprofit treated those with direct connection any differently.  As we’ve seen, customizing your message to your donor’s identity can further improve their retention.  Thus, identity is important to know at point of acquisition also.

In second place is commitment. Those donors with high commitment had a retention rate of 54% versus those with low commitment at 31%.  And, as with identity, customizing based on commitment level can create even more gains.  Donors with lower commitment need more initial communications to educate themselves about the organization and feel good about their choice, while already committed donors are turned off by these communications.

That puts satisfaction in first place.  Was there ever any doubt?  Those who said they liked their sign-up experience had a 68% retention rate; those who said they didn’t had a 16% retention rate.  Here again, those who were dissatisfied but were then communicated with appropriately to fix their problems saw their retention rates increase.

Here I should mention that the majority response to satisfaction questions was “no response” – because satisfaction must be answered after sign-up response rate will lag;  the other predictive variables can be gathered at sign-up.  The retention rate for those who didn’t answer was about half that of those who answered as satisfied and about double those who answered as dissatisfied – to be expected.

So, all four of these measures have their place:

  • Demographics for picking the potential donors to whom you most want to speak to
  • Identity and commitment for customizing communications
  • Satisfaction for sussing out problems in the process and fixing them both generally and for the dissatisfied donor

The best strategy is to combine all these factors into a model that predicts sustainer retention accurately.  While you will never have full second-sight, you can get to 80%+ accuracy at predicting retention, helping you pick the best donors to acquire and customize their journey once acquired.

Hope this helps you get and keep more sustaining donors!

Nick

P.S. You can learn more about this three-step process to increase retention in this Nonprofit Pro article or learn more about the sustainer retention platform here.

2 responses to “Predicting Sustainer Retention”

  1. Don Kossuth says:

    Thanks for sharing these insights, Nick.
    Two questions, if I may:
    When you say a group “had a retention rate of xx%”, are you referring to retention 12 months after sign-up, or …?
    And what type of completion rates are you seeing to sign-up surveying by email?

    • – This was actually a five-month retention rate for F2F donors (given that, I can’t emphasize enough that this was the “before”).

      – For F2F, commitment and identity are done at point of acquisition, so that’s 100%. Satisfaction is done after and thus is substantially lower: 7% is about the average; using best practices can get this up to 10%; our goal is generally about 15%. For other media, we’ve seen 7% response rates to no-ask surveys in mail and similar online.

      Since this is a “before” report, we couldn’t include it in the model, but people who don’t answer surveys also have a substantially lower retention rate (see for example https://agitator.thedonorvoice.com/what-about-people-who-dont-answer-donor-surveys/). Their lack of answer is itself an answer so we can learn from that absence.