Welcome to Conversion!

As of 2022, Widerfunnel and Conversion have officially merged under the Conversion name to form the world's premiere experimentation and UX firm.

Increasing Unicef's average one-off donation size by more than 50%

Client Unicef
Industry Not-for-profit

Unicef hired us to help them use experimentation to discover new ways of raising donations quickly. By running extensive user testing and applying insights from recent behavioral economics research, we were able to increase Unicef's average one-off donation size by 51%.


Unicef are one of the world’s leading not-for-profit organizations, providing humanitarian assistance in developing countries and saving millions of lives each year.

When we started working with them, they had an interesting dilemma:

In Unicef’s work, crisis is common – and reacting quickly to emergencies can be the difference between life and death.

At the time, they had a reliable flow of funds coming in from monthly donations, but they were looking for a way to raise money at short notice to help them better respond to unforeseen emergencies. Given the difficulty of scaling monthly donations quickly, they felt that the best way to do this would be to increase their average donation size.

That’s where we came in:

Unicef understood the power of experimentation, so they hired us to galvanise their donors and increase donation amounts.

Understanding the problem

From our user research, we uncovered a number of key insights that we believed could be used to drive the right donors towards a behavior switch.

Firstly, one of our earlier experiments indicated that Unicef’s donors gain more satisfaction when they know about the overarching cause that their donation is contributing to. When we started working on Unicef’s website, a lot of the messaging was focused around very specific goals that a donation would support. Things like the number of glasses of water a donation would buy. By focussing on the overarching cause rather than the particulars, we were confident we’d be able to influence user behavior in the desired direction.

Secondly, from the results of an earlier experiment, we knew that one-off donors were more likely to increase their donation than monthly donors. This suggested that we might have more success focussing our efforts on the one-off donor segment.

Finally, research from the field of behavioral economics showed that displaying a donation target on our donation page could increase donation amounts considerably. This would be particularly useful for Unicef if they found themselves in a situation where they needed to raise funds quickly.

Unfortunately, there was a catch:

Further research suggested that users were generally more motivated to donate when the progress percentage was higher –  so, for example, if a donation target showed as being 80% complete, people would be more likely to donate than if it showed as being 20% complete.

Our concern was that by displaying a lower progress percentage, we might actually reduce the number of donations Unicef received.

Pulling together to make a difference

Rallying people together around a single cause is a proven technique for generating results – the political world absolutely thrives off of it. Our strategy was to take this technique and apply it to Unicef’s website with the aim of increasing their total donations.

Based on all of the research above, we hypothesized that displaying a donation target alongside a target progress bar would produce an uplift in the average donation size and total donation amount.

Our primary concern, informed by the behavioral economics research cited above, was that a progress bar might disincentivise users from donating if the progress percentage was too low. One way around this would have been to display an artificially high progress percentage at all times, but Unicef were understandably eager not to mislead their donors by displaying a false progress report.

In the end we built a custom script to integrate the donation target display with Unicef’s google analytics account, allowing us to dynamically update the progress percentage and progress bar every time a new donation was received.

We chose to set the time period for the target as a week rather than a month, and hoped that by doing so, we would minimise any of the adverse effects that might arise as a result of displaying a lower progress percentage.


“Great to work with, and always looking to find ways of adding value to the partnership.”

James McDiarmid Head of Digital


This experiment increased the average size of single donations by 51%.

Digging deeper into the data, we found that all donor types donated more as a result of the changes, but that the increase in donation size was greatest for the >£100 category of donors. This was most likely due to a combination of the price anchoring effect and the economic ability of these higher donating individuals to donate even more. These were learnings that we were able to take forward and apply to other experiments to continue driving results for Unicef.

Ultimately, the experiment was a huge success, not only producing a tangible impact on average donation size and total donation amount but also yielding a number of key learnings – like the one above –  about Unicef’s users.

What’s more, by allowing them to increase their donation target at any time, we gave Unicef the means to quickly increase their flow of donations whenever the need arose. Given the unpredictability of crises around the world, the value of this capability can hardly be overstated.


Increase in average one-off donation size

Want to work with us?

We work with everyone from small startups to some of the biggest brands in the world.

Get in touch and see how we can help you.

Contact us