A Public Media Organization

Does concise donation form copy outperform long-form messaging?

Experiment ID: #199765

A Public Media Organization

Experiment Summary

Timeframe: 09/18/2025 - 10/06/2025

This experiment was conducted on WUNC’s primary donation form, which receives the majority of donor traffic from email, direct navigation, and on-site promotions. The goal was to evaluate whether shorter, more action-oriented copy would improve donor conversion rates compared to WUNC’s traditional, longer-form messaging.

WUNC’s standard donation form historically emphasized emotional storytelling, a sense of community, and shared responsibility for supporting public radio in North Carolina. However, recent research in donation behavior and user experience has suggested that minimizing friction—especially in digital donation flows—can lead to higher conversion rates. Internally, there was interest in exploring whether this applied to WUNC’s audience, especially during high-engagement periods.

The team hypothesized that short, punchy copy emphasizing urgency and clarity might increase conversion by reducing cognitive load and decision-making time. However, there was concern that cutting too much might sacrifice the warmth, trust, and shared identity that WUNC’s donors often cite as core motivations for giving.

Research Question

If we shorten and clarify the donation form copy to present a concise value proposition, then more donors will complete their gifts (conversion rate increase) and potentially give at higher levels (revenue increase). This is because WUNC’s multichannel donors are already primed and don’t need long explanations—clear, streamlined copy will reduce friction.

Design

C: Control

T1: Variation 1

Results

	Treatment Name	Conv. Rate	Relative Difference	Confidence	Average Gift
C:	Control	41.8%			$76.00
T1:	Variation 1	23.1%	-44.7%	100.0%	$72.00

This experiment has a required sample size of 48 in order to be valid. Since the experiment had a total sample size of 750, and the level of confidence is above 95% the experiment results are valid.

Flux Metrics Affected

The Flux Metrics analyze the three primary metrics that affect revenue (traffic, conversion rate, and average gift). This experiment produced the following results:

0% increase in traffic
× 44.7% decrease in conversion rate
× 5.3% decrease in average gift
47.6% decrease in revenue

Key Learnings

Main Takeaway

The long copy (control) outperformed the short copy (test) in total donations, first-time donors, sustaining gifts, and average gift amount. While the short version may have reduced friction, it lost the tone of community, empathy, and shared ownership that motivates many WUNC donors—especially first-time ones.

Performance Metrics and Insights:

Total Donations
Control (Long): 151
Test (Short): 90
Confidence Level: 99.99% – statistically significant
First-Time Donors
Control: 87
Test: 58
Confidence Level: 97.7% – statistically significant
Sustaining Gifts
Control: 80
Test: 48
Confidence Level: 98.9% – statistically significant
One-Time Gifts
Control: 71
Test: 42
Confidence Level: 96.5% – statistically significant

Observations:

The emotional, explanatory tone of the long copy more effectively supported WUNC’s value proposition, especially for first-time and sustaining donors.
The short version created clarity but may have diminished trust, emotional resonance, or perceived urgency.
Multichannel donors arriving to the form may have been primed—but still valued reaffirmation and community framing at the point of conversion.

Next Steps

Revise Short Copy Strategy: Test a revised short version that preserves the emotional tone, gratitude, and WUNC’s unique identity while still reducing friction.
Test with New Form Design: Consider testing updated copy with newer or simplified form layouts, where friction may already be lower.
Segment Messaging by Device or Campaign: Investigate whether short copy performs better on mobile or with acquisition-focused campaigns. Form copy may need to vary by context.

Experiment Documented by NextAfter

Question about experiment #199765

If you have any questions about this experiment or would like additional details not discussed above, please feel free to contact them directly.