Donor Behavior vs.AND Donor Attitudes

It seems the vast majority of non-profits define loyalty based on past behavior.  And like the financial, Cover Your Ass (CYA) language in all prospectuses reminds us, past behavior is no guarantee of future performance.

But neither of course is a purely attitudinal basis for segmenting and understanding your donor file because it is equally true that how someone feels or thinks never perfectly guarantees what that someone will do.

One can however, inform the other since every donor or constituent belongs in multiple segments – whether we know, measure and track it or not – and knowing provides more AND different information about the segment and its members.

Does anyone believe that a behavior based select for a house file renewal is so efficient that all the donors are attitudinally loyal – that is emotionally connected and committed to the group?

Ok, that was rhetorical but what about going further out on that limb and suggesting a large portion, half or more, of those folks receiving the renewal piece, despite being identified as “good” based on recent, past behavior, are NOT emotionally connected and committed?

And what if those non-committed have a drastically lower response rate and lifetime value?  So low that most are not financially worth mailing, even if you changed the economics of the package.

If you accept this is as true, at least for discussion sake, it is worth pointing out that an attitudinal segmentation is not likely to solve this problem for you, at least not initially.  This is because the weakness of most attitudinal segmentation schemes, since they are based on survey data, is that, unlike a behavior based scheme, you do not have segment membership for the entire file, only those who were sampled in the survey, which is typically a very small percentage of the whole file.    This means you can’t simply overlay your attitude segment onto your behavior one and only target those who, for example, are a “1” code on behavior (for “good”) and “2” code on attitude (also for “good”).  You simply won’t have the “2”’s for everybody.

However, and the details get beyond the purpose of this post, it is possible, with a large enough group of donors on your file tagged with an attitude segmentation code, to build a model to score the whole file.  This is typical stuff for the most innovative commercial operators.

But just because scoring the whole file is a ways off, non-profits need, we’d argue, to start somewhere and that somewhere is with getting more focused and serious about donor attitudes.  Roughly 80% of commercial entities in the Fortune 1000 use attitudinal segmentations to inform what they know and do about their customers.  We’d hazard that the percentage is closer to 20% for non-profits.

What can you do with attitudinal segmentations immediately?  You use it to dictate what you say and how you say it after using the behavior piece to figure out who you say it too.

The ‘what’ and ‘how’ of donor communication will seldom if ever be answered by a behavior based segmentation and NEVER as well as survey data (when done properly).

This means the two most important things you do in a direct marketing effort – selection and communication – require behavior and attitude insight to maximize efficiency and effectiveness.

Only using behavior to select and then relying on past experience, gut instinct or a myriad of often incomplete and contradictory advice on what and how to communicate is, in a word, crazy, considering how much of direct marketing performance is tied to the what and how pieces.

A properly conducted attitudinal study, whose purpose is to inform a behavior based view of the world and help determine what and how to communicate, is as necessary as the targeting piece and non-profits need to be better educated on the why and how of getting it done.

 P.S. Some might argue that direct mail testing, putting live results out into the market is the better option to attitudinal survey research since the former is “real world” and linked to results.

A fuller response on this will be forthcoming in another post but suffice to say, there are two main retorts here,

1)      Trying to discern what to say and how to say it at the segment level is strategic, or should be.  Testing of direct mail ideas is tactical and important but incredibly inefficient, costly guess work, to try and inform strategy.  We are not anti-testing but since the vast majority of tests do not beat the control, it is costing non-profits a lot of time and money to figure out how NOT to talk with donors.  It is time to figure out, strategically and systematically, what TO say.

2)      Properly conducted survey research CAN be linked to the bottom line.