For years, Domino’s believed it was in the transportation business. They spent a lot of operational time and money on speed, routing and driver density. Everything revolved around getting a box from oven to doorstep faster than the other guy.

Then fees rose, tipping fatigue set in, inflation crept up and customers turned carryout into a new preference.  The overlap between delivery loyalists and pickup loyalists turned out to be small, people weren’t toggling, they were behaviorally patterned.

We see the same thing in donor files.

Our analysis shows the overlap between check writers and online givers is in the single digits. The fuzzy middle barely exists.

Yet the default mail package still has a reply envelope and reply slip.  These aren’t neutral artifacts, they’re signals to how you think giving is supposed to work. For donors who no longer write checks, that signal creates friction before intention has a chance to convert.

This isn’t an argument against mail. Mail remains a powerful connector and priming environment, creating attention and narrative space.

But if someone is likely to complete digitally, the activation path should look digital. Not as an afterthought, not as a tiny URL buried under a reply slip, or a QR code slapped on but as the primary lane. Show the mobile form, make the wallet options explicit and signal speed and security in plain language.  The job to be done is removing ambiguity about what happens next by matching the dominant behavior pattern so “do it now” feels obvious.

And no, this is not guesswork. We model it using third party data to score acquisition files for likely payment preference the same way we score for Big Five trait profiles. The signals are not clean or labeled “this person uses Apple Pay.” Third party data is messy, it’s thousands of raw ingredients with no recipe card attached.

But people aren’t random and who we are in the world shows up in patterned ways. The media we consume, the devices we use, the subscriptions we hold, the way we transact in adjacent categories, all of it leaves traces. When you aggregate enough noisy signals, structure emerges and recipes become buildable.

The same stability shows up in timing.

Mode of 1 donors aren’t random annual givers. More than half give in the exact same month and if you expand the window to plus or minus that preferred month, you capture 90% or more of the giving. Yet the volume machine keeps pushing touches into the other nine months because activity feels prudent.

None of this is a critique of fundraisers. Most teams are operating inside inherited systems built on legacy signals, especially from an era when checks dominated and cooperative models were calibrated around check response. The system rewards motion and often undercounts digitally completed mail response unless matchback is handled with discipline.

A Human Operating System starts with a different question. What is this person’s pattern, and how do we design around it?

The donors are telling us who they are through their consistency. The only real question is whether our system is willing to listen.

Kevin