The problem begins, as usual, in a spreadsheet. You must budget direct marketing and your board will not accept something with three lines: revenue, expenses, and net.  How then to split your marketing programs?

Naturally, you do this by channel, since that’s how you’ll be spending money. A rational decision spurs irrational implications of thinking of digital donors, mail donors, telemarketing donors, as if it is the channel that defines the person who donates through that channel.  Pretty soon, it’s how your strategy is organized, zooming in on ever-smaller channel fractals (e.g., direct marketing to digital to social media to Facebook to Facebook Custom Audience to Facebook Custom Audience look-alike sectors).

All because of a @#)($*&@#()$ spreadsheet.  This would be fine if 1) channels were the best ways to organize donors, 2) strategies were best developed for channels, or 3) channel thinking were a benign influence.

None of these are the case.

Audience identity is a better way to organize donors than channels.  Ideal groupings of donors would have similar commitments to the organization, reasons for supporting the organization, and financial benefits.  Or, put another way, you want to segment to maximize the similarity of the people in the bucket each other and the difference from those outside.

Channel doesn’t do that.  Here’s an example from a health charity (some numbers tweaked for anonymity):

Pretty thin gruel here. What you might learn here are mail donors a bit more valuable and that people tend to stay with their own channel.  But when writing the appeal, you have no idea what to say for the mail group that is different from digital.

Now let’s look at this same audience, but by whether they were a direct beneficiary of services, indirect beneficiary, or no connection:

Now we have some things that are actionable.  Direct beneficiaries are the most valuable and most committed by a significant margin — there’s far more differentiation here than by channel.  They are also more likely to attend your events.  Each of these donors has a mailbox, email box, and phone number and it’s just a matter of their preference by which they give.

In fact, when you study these groups, you find that 60% of the relationship with direct connections is their involvement with the organization and patient care services.  This is literally zero percent of the relationship with those with no connection.

So not only do you know what group is most valuable, you also know what to say to them.  Yes, this takes a study of commitment and identities, but it’s well worth it for this type of intelligence.

Strategies are best done around audiences.  You can already see why — audience groups are far more different from each other than people giving via different channels (especially when the channel used to acquire someone can be an accident of history, not an actual preference).

But it’s also because channels are ephemeral. I was reminded of this by a piece written almost a decade ago called MySpace.com: A Place for Donors?

I should mention I am a fan of Care2-type services , so it’s nothing against them.  Also, in a decade much of what I write here will look hilarious.  For example, a couple of weeks ago, I referred to the rapper 50 Cent, which will look a bit odd in 2027 when he is an astronaut, felon, Vice President, or all three.

If I’m remembering 2007 correctly, there were also discussions of whether Second Life was a relevant channel for nonprofits and how much experience is necessary to become president.  It was a different time.

The reason that MySpace and Second Life came up then and Snapchat comes up now is that the channel is the new shiny.  What is constant is the audience.  Since a large part of strategy is determining what not do to, “how do I best reach this set of people?” will yield better fundraising results than “what do I do with this channel?” in the same way that “how do I build a bookcase” yields better home improvement results than “what fun thing can do with this hacksaw and welding set?”.

Channel thinking is actively destructive. Should you mail your event donors?  If you are letting channel drive, there are three ways to answer this question: 1) no, 2) yes, 3) yes, but take channel into account when you are doing your RFM selects so you aren’t going as deep into lapsed audiences.  The data would say don’t try too hard — most event donors don’t donate in other ways.

But consider if you did this by audience.  You would go deeper into direct connections than those with no connection to your organization and coming closer to the ideal mix for your program.

(And, in fact, both of these are not as good as the actual best answer: ask them.)

Channel thinking also hurts campaign structure: because different people are working on different programs, they may be deploying different messages to the same audience simultaneously.  This siloing has been well and roundly condemned in other places, but it leads to cases where, for example, your Google Grant keywords won’t include the subjects of campaigns you are running offline (or even in email) — you can’t benefit from the collective intelligence of everyone involved in your program.

So, let’s a fond farewell to channel thinking and bury it next to MySpace as a fundraising platform.