The Hypothesis
Ask arrays are one of the highest-leverage elements on a donation page. Not only do they suggest amounts to the donors to alleviate decision making, but they also anchor expectations.
We hypothesized that re-ordering the ask array from highest to lowest, combined with social proof (“Popular Giving Level”), would encourage donors to select higher gift amounts and increase overall revenue.
Specifically:
- A descending ask array would anchor donors to higher values.
- Highlighting a “popular” amount would create confidence and reduce friction.
- These changes together would lift average gift and overall revenue.
The Test Design
This optimization was conducted in two rounds, allowing us to test the theory, observe donor behavior, and refine responsibly.
Control
- Existing donation page
- Ask array ordered from lowest to highest
- $50 included as the lowest option
- No pre-selected amount or social proof
Variant (Round 1)
- Ask array reordered from highest to lowest (starting at $250)
- Lowest ask increased to $75
- $150 pre-selected and labeled “Popular Giving Level”
- All other page elements unchanged
Round 1 Results
Timeframe: May 20, 2025 – July 3, 2025
Key outcome:
The Control outperformed the Variant by 53.6%.
While statistical significance was not reached, the magnitude of the drop was impossible to ignore. We paused the test.
What Went Wrong (and Why That Matters)
Digging into historical giving data revealed a critical insight:
- Over 25% of donors gave between $50 and $74.99.
- The majority of those gave exactly $50.
By removing the $50 option, we unintentionally:
- Eliminated the most common donor choice
- Forced donors into amounts that felt unfamiliar or uncomfortable
- Created friction instead of guidance
This wasn’t “donor resistance to higher asks.” It was donors saying, “You removed the amount I came here to give.”
That’s a classic CRO mistake, and it’s exactly why testing matters.
Round 2: Refining the Test
Rather than abandoning the concept, we isolated the variable.
Variant (Round 2)
- Descending ask array retained
- $50 reinstated
- $150 still pre-selected as “Popular Giving Level”
- No increase to the lowest amount
Round 2 Results
Timeframe: September 10, 2025 – October 29, 2025
Key takeaway:
Conversion rates were statistically similar, but the Variant produced higher total revenue and a lift in average gift.
Final Conclusions
Three clear lessons emerged:
- The $50 ask is important. Removing it materially harmed performance.
- Descending ask arrays can increase average gift—without killing conversion—if donor comfort is preserved.
- Social proof works best when it guides, not coerces. “Popular Giving Level” performed best when paired with familiar options.
Importantly, this test reaffirmed something we emphasize often: Optimization isn’t about forcing outcomes. It’s about learning donor behavior quickly and responding intelligently.
What’s Next
Future tests will explore:
- Increasing the top anchor (e.g., $500 or $1,000) while keeping $50 intact
- Testing “Popular” placement without pre-selection
- Segmenting ask arrays by traffic source or donor status
The donation page is already performing within industry benchmarks. These tests are about incremental gains, not fixing something broken.
