How the Greater Vancouver Food Bank Turned Email Into a Million-Dollar Channel
February 12, 2026
Email had always been an add-on to their direct mail. We treated it like the revenue channel it could be, and it now raises over $1 million online a year.
Increase in email revenue from the Generosity X program.
Higher giving rate among at-risk LYBUNTY donors, at 99.99% probability.
Decrease in unsubscribe rate, even at twice the email volume.
The challenge
Email was an add-on to direct mail, not a real channel.
The Greater Vancouver Food Bank built strong digital fundraising during the pandemic, then watched online revenue spike and plateau. Within that period, email never became a meaningful part of their fundraising. It was along for the ride, a light add-on that the Food Bank’s direct mail agency ran as an extension of the mail, never built as a channel in its own right.
Many organizations saw the same pandemic surge and the plateau that followed, and the fact that the Food Bank’s online revenue has since bounced back is part of what makes this story worth telling.
In our experience that gap is a big opportunity left on the table. A focused email program can raise millions for an organization this size, but only if it is treated as more than an afterthought. The usual instinct is to send less email, not more, for fear of wearing donors out. That assumption rarely gets tested, so it quietly caps what the channel can raise.
The Food Bank needed a partner who could:
- 1Treat email as a real revenue channel, not an add-on to the mail.
- 2Test a real change instead of guessing at what email could do.
- 3Grow revenue without burning down donor goodwill.
Our approach
We didn’t guess. We ran a controlled experiment.
We split the file into two groups. The control group kept the existing email cadence, the version the incumbent direct mail agency ran alongside the mail. The Generosity X group received a dedicated program of our own design, and we ran both across the key December giving window.
Testing against the cadence already built with the mail program is a fair, honest comparison. We wanted to know whether a dedicated email approach could outperform what was already in place, not to take anything away from the mail.
Then, instead of eyeballing totals, we tested the results for statistical significance, asking whether each difference was real or just noise. We reported what held up and were transparent about the confidence behind each result, because rigor is the entire point. That is the difference between a real finding and a lucky month.
More touchpoints with purpose
About twice the email volume, balanced across asks, thank-yous, and impact updates.
Human-centric messaging
Mostly plain text, written to feel like a note from one person to another.
Segmentation and personalization
Messages tailored to mid-level, monthly, and at-risk donors so each one stayed relevant.
The experiment
What the test actually moved.
The full set of results, control group versus the Generosity X program. For the two audiences we tested for significance, open The statistics tab to see the probability model behind the win. Hover or tap any bar for detail.
How to read the bars
Every result here compares a control group, the existing approach, against the Generosity X treatment. We set the control to 100 as a baseline, then measure the Generosity X group against it. A bar at 200 is twice the control, and a bar below 100 sits under it. Indexing to a shared 100 puts each result on one like-for-like scale, so you can weigh the size of a lift on its own terms, separate from the raw dollars or counts behind it.
The result
More email, more revenue, and donors who stayed.
The Generosity X group more than doubled email revenue and lifted total donations across all channels by 11 percent. New gift acquisition rose by 4.24 percent, with a 92 percent probability the gain was real and not chance. Among LYBUNTY donors, those who gave last year but not yet this year, the giving rate climbed by 60 percent, at 99.99 percent probability. Unsubscribes did not climb. They fell by 27 percent, even at twice the volume.
