Response First, Value Intact: A House Mailing Test

What?

“At a medium-sized nonprofit, the direct mail team ran an in-house test on 30,000 donors. They aimed the prediction model at response—not gift size—to keep more people active. The tradeoff showed up fast: response rose while average donation fell. Even so, revenue per letter climbed, and profit per letter increased. The uplift was statistically significant, proving that response-first can support file growth without sinking economics.”

Response rate is the share of mailed donors who give; profit per letter counts income minus costs, divided by letters.

So What?

For fundraisers who must keep the file active, response matters. This test shows you can focus on response and still improve unit economics—profit per letter rose—while feeding acquisition and reactivation goals.

Insight: you can optimize for response, ROI, or a balanced objective. Even a response-maximizing setup can deliver positive ROI.

Now What?

Try this sequence this week:

1. Score for likelihood to respond; build mailing bands around the top scorers.
2. Hold out a control and track revenue and profit per letter, not response alone.
3. After the first wave, re-target only responders with upgrade asks to recover average gift.

Results

• Representative test; 30k target size; uplift statistically significant.
• Model optimized for response; detailed percentages summarized in highlights above.

Key Takeaways

• Response-first can still lift profit.
• Track profit per letter, not response alone.
• Shift toward balance if gift size dips.

Curiosity Spark

Where would you set your first “Predict → Mail → Measure → Balance” threshold?