Is Your Online Donation Form Turning Off Donors? | npENGAGE

Is Your Online Donation Form Turning Off Donors?

By on May 3, 2018


online donations

If there is one thing that is almost entirely unique to the world of nonprofit websites (compared to the for-profit companies I have assisted for the past 10 years), it is the part of the donation form (or response device) we call the gift array.

This is also referred to as an ask array, an ask ladder, or suggested gift amounts. This assembling of giving options (or buttons, depending on the site) seems to be a common staple on just about everybody’s main online donation form.

In the for-profit world, we really only see this in buying gift cards. And though a gift is about to happen, the situation is different as it is expected that there will be some sort of product/service that comes with that.

So here’s my question: has anyone really asked why we put them there in the first place? Why not just let them tell us what they want to give? Chances are, we put them there because someone once said that they inspire people to be more generous, or to give when they wouldn’t have. And actually, that may well have been true for a given situation and time.

Does this gift array, and its presentation, really have any impact on generosity RIGHT NOW?  Is it OK that people are just defaulting to gift arrays? And is it OK that they are starting with a higher amount rather than a lower amount?

At NextAfter, we’re in the middle of conducting a field study on how organizations respond to certain giving situations, and part of that study involved us making a massive amount of anonymous donations on main donation pages across just about every different nonprofit vertical. And what we noticed is that a) organizations are all about using the classic gift array and b) a LOT of organizations like to start with HIGHER amounts first in the eye-path.

Here is an example (left to right)

website donation form

Or a mobile example (top to bottom)

Mobile donation form

I mean really, why would you start with a higher amount? What is the logic?

“Well, it’s going to encourage people to give more,” someone thought, “because higher amounts are presented first as their options…and by emphasizing giving more we will convince them to give more!”

Is that how we really think of our donors?

The real question we need to ask is this: what effect does the gift array’s presentation truly have on individuals who are contemplating that gift? Does it really affect them?

Obviously some people will not care and make their donations regardless. But if there is a large enough group of people that DO care, I wouldn’t want to lose them at this donation opportunity (and possibly forever) because of my array presentation.

So we put it to the test. One of our partners, CaringBridge, offers free personal, protected websites for people to easily share updates and receive support and encouragement from their community during a health journey. Here is what their gift array on most of their donation pages looks like:


The control (original) uses a rising suggested amount approach (start with the lowest ($50) on the left (assuming visitors in this case naturally read left to right) and then end with the highest ($250). On mobile, it stacks on top of each other with the lower amount first.

We put this approach to the test by switching the $50 and $250 options, so that people reading left to right would see the HIGHER option FIRST. And the same in mobile… higher amount first stacked.


donation gift array


So what was the result?

The treatment’s high-to-low emphasis approach achieved a whopping 15.7% DECREASE in gifts, AND a 11.3% DECREASE in Average Gift size, resulting in a total 25.2% statistically significant DECREASE in giving revenue.

So, by showing the larger amount first, many visitors were LESS likely to donate, and LESS likely to give in larger amounts. And that was the ONLY change. Nothing else.

But why?

“We have found that people give to not-for-profits not as faceless organizations but humanize them as people…”  -Josh McQueen

It is possible that people see your array as more than just an array.  They also see the way in which you present the array as a point of communication from you, much like how body language communicates in real life. If this is the case, then let’s examine what this high-to-low approach subtly communicates to someone:

  • Lower amounts are less acceptable.
  • “Sure, we’ll TAKE your donation, but we might not appear as happy about it, or, we really don’t prefer the lower amounts… that’s why they’re last… duh.”

This would explain the drop in completed gifts altogether.  Some people (to the tune of 15.7%) probably felt that their small gift wouldn’t be appreciated, simply because it was at the bottom of the list. It would be no different if the gift array started at $500 and moved its way up the ladder to $5,000.

“Anything less than $500 isn’t significant,” the array says. Even though any gift is better than zero, that is not communicated here. When one starts backwards, from high to low, the lower gifts are devalued psychologically.

But wait… shouldn’t the average gift amount have at least gone up??? After all, we are emphasizing larger gifts, so even if we lose some donations, the larger gifts should have made up for it. That’s the fundraiser’s mindset, not the donor’s. To that stereotypical fundraiser, it’s moneyball, statistics, number crunching… winning the game. To the actual person giving the gift, it’s something entirely different.

Let’s think about the donor that wants to give $50. When the array is presented high to low, then $50 ALREADY looks bad. If the donor upgrades to $60… $75…what difference does that make? According to the high-to-low array, the organization doesn’t really notice. They notice the big gifts, so there is no additional benefit to the giver to give a little more because it seems like the organization doesn’t want it or care. BUT… what about the array that goes from low to high?

So back to the $50 giver. The first option they see is the lowest – $50. And they are thinking… “you know, I really appreciate this organization so much… how about I give a little more?” And all the sudden that increase in giving becomes an UPGRADE.  

Now the donor feels like their slight donation increase just morphed into a mid-tier gift, instead of being an “unappreciated low tier gift” despite the increase in amount, because it rose above what appears to be what the organization deems as desirable (meaning, we as humans interpret their first option as what is considered desirable and acceptable, similar to how we interpret a body that leans into a conversation as interested).


“Users [of digital experiences] will infer a psychology whether or not the designers intended this. For this reason, I believe designers must embed appropriate psychological cues.”  -Dr. B.J. Fogg

Your donation page carries a conversation with the person whether you like it or not. People read into this stuff!

It is our responsibility to make sure our digital experiences communicate how we truly feel about our donors: deep appreciation. And if we truly don’t want to accept someone’s lower donation amount and only want donors willing to give a minimum large amount, then we get what we deserve.

Read the original experiment write up from NextAfter here.


As Senior Director of Research and Education, Jon Powell is wholly focused on taking everything the NextAfter team is learning and transforming it into insightful, practical and immediately actionable advice for marketers and fundraisers, regardless of their organization size.

Jon knows firsthand the challenges marketers face: he has experience building an entire digital marketing department from scratch as Director of Digital Marketing at B+B SmartWorx and has more than eight years of hands-on marketing optimization experience gained through managing hundreds of A/B and multivariate tests at the MECLABS Institute.

In addition, Jon has already conducted multiple in-depth meta-analyses of the thousands of case studies that are held in the research library of MECLABS Institute, one of the largest independent databases of experiments for marketing and sales in the world.

Comments (5)

  • Lisa says:

    Was this only tested on one site or was it tested on multiple sites? If only on one site, it’s hard to make these statements as it could just be specific to donors on this one site. I would love to see if the assumptions hold true across multiple donation pages on organizational websites.

  • Lauren says:

    I would be curious as well to see if this was tested on multiple websites. Or if any other users have changed their gift array and seen an increase/decrease

  • Joyce says:

    I was donating on a site recently and noticed that I had the same psychological response that you are describing, where I felt that the amount that I was intending to give wasn’t going to be appreciated because it was less than the “desired” donation amount . So instead of feeling “good” about my donation and generosity, I felt “bad “ that it wasn’t enough. I noticed that my take away emotion was not that positive , and likely affects my desire to donate to that cause in the future.

  • Claire Axelrad says:

    You’re making a lot of inferences from just one test listing just three suggested donation amounts. Might the results have been different if there were five suggestions? If the amounts were closer together in size? Perhaps. I don’t think you can generalize that using high – low ask strings is a bad idea across the board from a single test.

    In fact, there’s plenty of research showing the positive “anchoring” effect of suggesting a higher number first. In one famous study researchers presented a beer menu to customers based on price in two ways: from low to high and from high to low. The latter resulted in an average of $0.24 more every time someone ordered a beer — a 4% increase. “Anchoring” suggests numbers act as psychological magnets.

    • Jon Powell says:

      Hi Claire, (and others!)

      Let me clarify that this article represents more of an abstract than a full data set. Of course I have more data points 🙂

      I actually have conducted a meta-analysis on gift arrays and their presentations thereof, with statistically significant data sets spanning 38 primary data experiments, as well as a review of multiple non-profit focused academic studies, not to mention multiple choice-set optimization based peer-reviewed academic articles that talk about the effect of choice in for-profit situations.

      What I’ve discovered is that when you look at it in the lens of “how much more can I persuade them” to give, you often get studies where the results contradict each other. For every study that says “this works”, I found one where it says “it doesn’t.”

      And especially if you bring for-profit studies in the mix (never heard of an organization giving away beer for the better of the world, but man… that would be awesome!), you’ll find that there are differences in results there as well.

      What I have discovered is that when you look at the data sets from a different point of view… meaning… how does this experience affect the human in this particular scenario… and when you actually consider the context they are in, or the source, or the environment, you actually start to see real patterns.

      That is why I have found at least 3 donation page types. It was the gift array study I conducted a year ago that showed me that… because only when I separated the data by source/context, I found patterns.

      This is why one cannot understand Shakespeare’s works fully without looking at the time period it was written for. Context is everything.

      What is the context of your data?

      So if you have data that fits the situation (non-profit), and also includes the context of the humans going through it (source, situation), I would absolutely love to add those findings to my data set. I have adjusted my conclusions over the years in certain areas (including email) as a result of that.

      And bear in mind… in every meta-analysis I’ve ever conducted, which is a lot of them (minimum of 30 experiments per meta… so millions of impressions..), I’ve never once found a single tactic that makes the same amount of difference on one audience compared to the next.

      What I do find that’s useful and reliable is that certain tactics have risk levels… and I recommend to others in good faith according to the data the tactics with risk levels that are extremely low based on all the data

      And then when I write about it, I write in such a way that helps them feel what I feel when I optimize pages and get wins based on the data I have in front of me. It’s entirely possible that you or anyone could interpret it slightly different. At the end of the day, though, does the interpretation I provide help someone get a win while also treating their prospect better? (I don’t believe in dirty tactics for the sake of squeezing a little more out of someone…)

      As for the tactic of starting high on your gift arrays? All the data I have suggests there is signficant risk present and that it should be avoided or handled cautiously at bear. (Meaning, optimize another part of your donation page before trying to optimize this)

      Once I get even more data to say otherwise (I would need a lot…), then I will stand-down the yellow alert. 🙂

      Does this help?

Leave a Reply

Your email address will not be published. Required fields are marked *