October closed at $12,001 raised from email — 16% of the $75,000 goal — with 98% of the donor list already solicited using the strongest psychological lever available. The math was unambiguous: even a perfect repeat of October’s performance in December would leave a $27,000+ gap. This is the story of the second channel we built to close it, and what the data said when we tried.
Table of Contents
By October 31st, the campaign had raised $12,001 from email — a 2.2% conversion rate achieved using the most powerful psychological lever available: a matching grant narrative with a hard deadline. 98% of the 3,001-person list had now seen that lever and chosen not to give. The narrative couldn’t be reused, ethically or practically. The question for November and December wasn’t how to run a better email campaign. It was whether email could close the remaining gap at all.
The arithmetic was straightforward and unforgiving. Even matching October’s 2.2% conversion rate again in December — without the matching grant narrative, against a list that had already said no once — would produce roughly the same $12,000. Combined with October’s results, that left the campaign tens of thousands of dollars short of $75,000. Mass email had a mathematical ceiling, and the campaign was already pressed against it.
The conclusion
The revised strategy split the remaining eight weeks into two simultaneous tracks, run in parallel rather than in sequence: personal cultivation of high-capacity donors, and a refined December email sequence built directly on what October’s A/B tests had already proven.
Track 1: Major Donor Cultivation
Roughly 50 high-capacity prospects — past donors of $500 or more, business connections, and community leaders — were identified for personal outreach: handwritten notes, phone calls, and one-on-one meetings, rather than mass email. The logic was simple. A handful of personal conversations with high-capacity individuals could deliver more revenue than thousands of additional email opens, and email had already shown it could not move this donor base further without a new psychological lever.
Track 2: A Refined December Email Sequence
Only one of October’s three A/B tests produced a confirmed result: urgency-led subject lines beat social proof by 7.9% on open rate, a finding applied to every December send. The other two tests stayed inconclusive. Email length (Test 2) showed identical 0.7% click rates between short and long versions — format had no measurable effect on conversion, so December’s emails weren’t built around a “short wins” rule that the data never actually established. Send time (Test 3) revealed a genuine trade-off rather than a winner: mornings drove more opens, evenings drove more clicks. December’s emails went out in the morning as a directional choice favoring reach, not because morning sends were proven superior — the underlying data still supports either timing depending on the objective.
Segmentation worked differently than in October. Rather than excluding donors from later sends, donors stayed in the sequence throughout Email #1 through #3, but received the donor-tailored version of each — lighter in urgency, framed around progress and gratitude rather than a first ask — while non-donors received the version built to convert. Where October treated the full list as one audience, December split it from the very first send: one version for October donors, one for non-donors, acknowledging that these groups now had a different relationship with the campaign. Only Email #4 and #5, in the campaign’s final stretch, narrowed to non-donors exclusively.
Giving Tuesday and the December 31st tax deadline were treated as milestone markers rather than a repeat of October’s matching-grant urgency — a deliberate choice to avoid signaling that “special offers” were a recurring negotiating tactic, which risked eroding the trust the matching narrative had spent.
November carried no mass fundraising emails. That absence was a decision, not a gap. The list had just been through six increasingly urgent emails in five weeks; sending a seventh in early November risked donor fatigue and unsubscribes without a new lever to justify it. Instead, November’s energy went entirely into two things: launching major donor outreach, and preparing the December sequence using October’s findings.
No email-driven revenue was expected or generated in November by design. The single transaction recorded in this window was the SVCF grant, arriving through the foundation relationship rather than any campaign send.
Five emails went out between December 2nd and December 31st — one fewer than the six originally planned, since the contingency final-day email was never triggered. Each send applied the one confirmed finding from October’s testing — urgency-led subject lines over social proof — along with morning timing as a directional choice and donor segmentation by version, wherever the audience size justified the split.
A note on revenue tracking
Without per-email revenue data, the clearest honest signal in December’s $18,830 is gift size. Treating $1,000 as the threshold for a major gift — a figure large enough to signal a relationship beyond a single email click — produces a sharp, defensible split.
of December’s $18,830 came from just 5 gifts of $1,000 or more — the Next Step Fund’s $10,000 personal-outreach gift plus four individual $1,000 gifts, totaling $14,000.
of December’s total — $4,830 — came from 30 gifts under $1,000, spread across the full email-reached donor base.
The same pattern held throughout the campaign — a small number of large gifts carrying a disproportionate share of revenue, while a much larger group of smaller gifts contributed the remainder. The full breakdown across all three phases, along with what it means for the campaign as a whole, is covered in Final Results & Key Learnings.
This section exists because two things are true at once: the data supports some attribution claims clearly, and it doesn’t support others, and the difference matters.
What the data supports: Email performance metrics — delivered counts, open rates, click rates — are fully verified from Network for Good’s email dashboard for every send in this sequence. Major donor cultivation and the December email campaign ran simultaneously throughout the same period, and gifts arriving via check, ACH transfer, or wire (such as the Next Step Fund’s $10,000 gift) are confirmed as arriving through channels with no connection to an email click.
What the data doesn’t support: a clean, per-email breakdown of December revenue. The list-group tags that would have shown which email a given December donation followed were deleted from Network for Good before they could be exported or preserved — a data-management error made while cleaning up old campaign lists, not a platform limitation. Once that happened, it became impossible to say with confidence that a particular gift was a direct response to Email #2 versus Email #3, or to separate a credit-card gift that followed an email click from one that arrived through a personal phone call where the donor simply chose to pay online.
The Lesson
Given that, this case study presents December’s revenue by gift size rather than by claimed channel — a methodology built entirely on verified transaction amounts, with no assumptions about donor intent or contact history required.
The strategic bet made in November — that closing a $27,000+ gap required a second channel, not a better email — held up under real execution. Five emails reaching thousands of non-donors produced steady, modest engagement; five major gifts produced nearly three-quarters of December’s revenue. Mass email had already proven its ceiling in October. December confirmed that the ceiling doesn’t move much, no matter how well the emails are optimized, once a list has been through its strongest psychological lever and said no.
What this meant for the campaign’s final outcome — and what carried forward into how future campaigns are planned — is covered next.
