By Maja Adena and Anselm Hager
Online fundraising is on the rise. Until recently, however, little was known about whether advertisements on Facebook or Google are actually effective at generating donations. To fill this gap, BCCP Senior Fellow Maja Adena and co-author Anselm Hager conducted a live experiment on Facebook with the NGO Save the Children. Together with the NGO, the researchers placed video ads on Facebook. To assess whether the ads work, the authors randomly assigned some zip-codes to be targeted by a 14-day ad campaign, while other zip codes did not see any ads. After the experiment, the researchers compared the actual volume of donations in the treatment and control zip-codes. The results show that the ads raised donations by almost €4 per €1 spent (on advertising) in treated zip codes. This constitutes a 300 percent return on the money invested.
The assessment of whether an internet advertising campaign works or does not is not as trivial as it seems. Advertisers are often easily misled by the level of direct response to an ad. On the one hand, if they observe donations through the ad link, this does not necessarily mean that there are more donations overall. Those donations might be coming from existing donors who find it convenient to use an online-link instead of their usual donation channel. On the other hand, if advertisers observe a low volume of donations through the ad link relative to the costs, it does not necessarily mean that the advertising is ineffective. Ad recipients might react with a lag, after some deliberation or after another trigger.
The authors overcome the above challenges to causally assess the effects of an online fundraising campaign with a smart research design—they observe donations from all channels over a longer period of time and compare donation volume in treated versus untreated zip codes. They conclude that firstly, the videofundraising campaign on Facebook increased the volume and frequency of charitable giving. Secondly, this holds true for short- as well as long-term donations, the latter pointing to a long-term advertising effect. Thirdly, they found that existing donors do not switch from offline to online giving and do not expedite their decision to donate after seeing online advertising. Online fundraising generates genuinely new donations. Fourthly, however, they also found that the fundraising campaign for Save the Children reduced donations to other similar charities, indicating that charities compete for scarce resources.
The full paper “Does online fundraising increase charitable giving? A nation-wide field experiment on Facebook” is available as WZB Discussion Paper SP II 2020–302.