Why a posted goal gets people to give more: what the research says

The thermometer drawn on a big poster near the school entrance or in the arena lobby is part of the fundraising landscape. Online, it becomes a progress bar. People use it because “it motivates.” Research lets us go a bit further: several studies have measured what happens when a goal is posted, when early gifts are announced and when a campaign gets close to its target.

This article summarizes those studies with their numbers, their limits and what they don’t say. The second half is practical: how to choose your number, how to display it and when to announce the first gifts.

What the studies measured

The closer the goal, the faster people move

The starting idea comes from psychology: the closer you are to a goal, the more effort you put into reaching it. In 2006, Ran Kivetz, Oleg Urminsky and Yuhuang Zheng tested it in a café at Columbia University with a loyalty card: buy ten coffees, get one free. Analyzing 949 completed cards, they saw that customers left less and less time between purchases as the card filled up. The time between two coffees shrank by about 20% from the start of the card to the end.

Their second test is more telling for a fundraiser. Some customers got a card with 10 empty boxes. Others got a card with 12 boxes, two of them already stamped as a “bonus.” In both cases, customers had to buy ten coffees. Customers who started with two boxes filled finished their card in 12.7 days on average, compared with 15.6 days for the others, almost three days sooner. Simply seeing that they were already on their way was enough.

These were coffee purchases. This study served as a starting point for researchers who later tested the same idea on generosity.

Gifts announced before the launch

In 1999, John List and David Lucking-Reiley ran a real campaign for a research centre at the University of Central Florida. They wrote to 3,000 households in the area, randomly split into six groups of 500. Each group was asked to fund a US$3,000 computer. The only difference between the letters was the share already raised: 10%, 33% or 67% of the cost.

The results are clear. With 10% already in hand, 3.7% of households gave. With 67%, 8.2% gave. The average gift also rose, from about $12 to about $35 in the groups without refunds. In total, the two 67% groups brought in $3,260, compared with $581 for the two 10% groups. The authors describe an increase of nearly six times. Promising to refund donors if the goal was not reached also helped, but much less: about 21% more.

Keep the context in mind. The targeted households earned more than $70,000 a year and had already given to a charity. The gifts were small (the largest was $250). And it was a single campaign, in a single city.

An announced lead gift, compared with a lower goal

One question remains after the previous study: is it the lead gift that matters, or simply the fact that there is less money left to find? Daniel Rondeau and John List tested this with the British Columbia chapter of the Sierra Club of Canada. They wrote to 3,000 supporters to fund an environmental education program in schools.

Four groups received different letters:

• $5,000 goal, nothing announced: $1,375 raised;
• $2,500 goal, nothing announced: $945;
• $5,000 goal with $2,500 already pledged by a lead donor: $1,620;
• $5,000 goal with a donor matching every gift: $1,235.

Announcing the lead gift therefore did better than lowering the goal by the same amount: 71% more than the $2,500 group, with the average gift going from $27.79 to $43.78. Matching gifts did no better than the other groups. One important limit: none of the four groups reached its goal, and response rates were much lower than usual for this organization.

People give more when the goal is close

In 2013, Cynthia Cryder, George Loewenstein and Howard Seltman published three studies on what they call the goal gradient in helping behaviour.

The first looked at the microloan website Kiva, where each request shows a progress bar. Across 209 requests tracked hour by hour for about a week, funding moved forward by 6.7% per hour when a request was between 0 and 33% of its goal, by 10.8% between 33 and 66%, and by 12.8% above 66%.

The second was a real mail appeal to 13,500 past donors of a disaster relief organization. Saying the project was at 10% or 66% of its goal changed nothing. Saying it was at 85% raised the giving rate from 0.5% to 1.17%, more than double. The average gift, however, did not move: the announcement mainly won over people who would otherwise have given nothing.

The third study presented 183 adults with the story of a Grade 7 student who has to sell 100 chocolate bars for her sports team. She had either 2 or 32 bars left to sell, and respondents were told she would reach her goal either way. Respondents said they were more inclined to buy when only two bars were left. Their explanation: near the end, each gesture seems to have more effect, and that feeling makes giving more satisfying.

What online platforms show

Rick Wash analyzed 344,196 classroom projects posted on DonorsChoose between 2007 and 2012, a site where American teachers fund school supplies. The gift that completes a project is on average much larger than the others ($207 compared with $36.54). After accounting for project size and other factors, the gap remains $74 to $93. People whose first gift completed a project also came back to give much more often: 55%, compared with 20% for the others. Finally, few projects stop just short of their goal.

Venkat Kuppuswamy and Barry Bayus studied the same dynamic on Kickstarter. Support grows as a project nears its goal, then drops off sharply once the goal is reached.

Ethan Mollick, who analyzed more than 48,500 Kickstarter projects, sums up the picture this way: projects succeed narrowly or fail by a wide margin. Failed projects collect on average 10.3% of their goal. A quarter of successful projects exceed their goal by no more than 3%. He also notes that the higher the goal, the less likely the project is to succeed.

What these studies don’t say

Before applying these results to your committee, a few cautions:

• None of these studies looks at a school or club in Quebec or elsewhere in Canada. The settings are American, except for the Sierra Club study, done in British Columbia.
• The thermometer poster itself has not been tested. The studies measure the effect of progress information: a percentage in a letter, a bar on a website. The poster conveys the same information, but nobody has compared a poster with no poster.
• The effects are real but modest in absolute numbers. Doubling a giving rate of 0.5% gives you 1.17%. That matters across thousands of letters. On a list of 80 families, it may mean one or two gifts.
• The beginning is the weak moment. Announcing that you are at 10% did nothing in Cryder’s study. A nearly empty thermometer probably doesn’t help much.

Choosing your goal

The studies pull in two directions. In Rondeau and List, the higher goal brought in more than the lower one. The authors themselves add that a huge goal would probably discourage people. In Mollick, big goals succeed less often. The right number is therefore high enough to look serious and low enough to be reached.

A simple method for a volunteer committee:

1. Start from a concrete expense. A bus for the end-of-year outing, jerseys for 18 players, books for the library. All the studies above involve a specific item (a computer, a program, a classroom project).
2. Look at what you actually raised last year, for the same activity and with roughly the same number of families. That is your best indicator of what is realistic.
3. Add a reasonable margin if you have more participants or better organization this year. Avoid doubling the goal without a reason.
4. Break a big project into stages. Wash’s study suggests that projects people can complete attract end-of-race gifts. A $12,000 goal can become three $4,000 stages, each tied to a purchase.
5. Plan what comes next if you pass the goal. Kuppuswamy and Bayus observe a drop in giving once the goal is reached. Say in advance what the surplus will pay for, or announce a second stage.

Announcing the first gifts before the launch

The studies by List and Lucking-Reiley and by Rondeau and List point the same way: announcing that part of the money is already in hand helps both participation and gift size.

In practice:

• Collect some first gifts before announcing the campaign to the public. The school’s governing board, the coaches, a few committed parents, a neighbourhood business. List and Lucking-Reiley cite a professional handbook that recommends having 40 to 50% of the goal in hand before the public phase. That figure comes from practitioners and was not measured: their study only shows that 67% did better than 33%, which did better than 10%.
• Announce those gifts clearly in the very first message. One sentence is enough: “Thanks to three local sponsors, $1,500 of the $4,000 is already raised.”
• A simple lead gift is as good as a match. In the Sierra Club study, announcing a lead gift worked better than a donor matching every gift. If your sponsor is torn between the two formats, nothing requires you to pick the match.
• Keep something in reserve for the end. If a sponsor agrees to complete the last 10%, announce it when the campaign is close to its goal. That is when progress information had the most effect in Cryder’s work.

Showing progress

The poster at the school or arena

• Put it where people walk by often: the entrance, the daycare bulletin board, near the cafeteria.
• Show three things: the goal in dollars, the amount raised, what the money will pay for.
• Don’t put it up empty. Wait until you have your first gifts so it starts with a good share already filled, like the two bonus boxes on the coffee card.
• Update it on a fixed day (every Monday, for example), and show the date of the last update.

The online progress bar

• Show the percentage and the amount left. In Cryder’s study, it was being close to the goal that moved people. “Only $600 to go” says more than “85%.”
• When you get close to the goal, say so in an email or a post. One sentence with the amount left and a deadline.
• Once the goal is reached, say thank you, then announce the next stage if you are continuing.

Checklist before launching

To check	Done
The goal is tied to a specific purchase or project	☐
The number compares with what you raised last year	☐
Part of the goal is already pledged before the public announcement	☐
The first message announces that amount already raised	☐
The poster or bar shows the goal, the amount raised and the update date	☐
A message is planned for when the campaign reaches about 80 to 85%	☐
You know what to say if the goal is exceeded	☐

In short

• Announcing that a good share of the goal was already raised clearly increased giving in two real experiments, in the United States and in British Columbia.
• Progress information helps mostly near the end: saying you are at 85% doubled the giving rate, while saying you are at 10% or 66% changed nothing.
• Goals that are too high fail more often: start from last year’s results and a concrete purchase.
• Don’t launch an empty poster. Collect your first gifts beforehand, and save a message for when the goal is close.
• No study has tested the thermometer poster itself. What we know concerns the information it conveys.