Self-donations in P2P fundraising

A self-donation is the best indicator of success for an individual P2P fundraiser. If a person donates themself, they are 3.7 times more likely to reach their fundraising goal than if they don’t. In this blog post we will explore everything there is to self-donation and P2P fundraising.

The positive effect on reaching your goal

34% of fundraisers who donate themselves reach their goal. It is just 12.2% of those who don’t donate themselves that reach their goal. This is a striking difference and indicates that self donations are tremendously important in P2P fundraising.   

The above is based on data on over 33.000 fundraisers. And if we do a little bit more calculation we can calculate the ratios between those who reach their goal and for both those who self donate and those who don’t. 

It is the ratio between these two ratios that give us the odds likelihood ratios of 3.7. In other words, if a P2P fundraiser donates to their fundraising page after creation, they are in other words 3.7 times (or 3.66 times to be precise) times more likely to reach their goal. 

There is thus no doubt that there is a strong correlation between donating yourself and your likelihood of reaching your fundraising goal. This is the reason that we recently launched a couple of new features to further encourage them.

Let’s explore why this effect exists, - why do self donations have a positive influence on the overall fundraising results?

Signalling

Let’s say your friend has created a fundraising page for a cause. You go to their fundraising page, but notice that your friend hasn't donated? Weird right? Your friend wants you to donate, but doesn’t themself? 

A fundraising page is comparable to your friend telling you about a new product you should try out, And not making a self-donation is (kinda) comparable to your friend not even having tried the product they recommend themself!

In a different scenario, your friend has donated themself - this is a different situation, and instantly creates more trust. Your friend not only shares and promotes this cause, but they also support it actively with their wallet.

From this, it should be obvious that self-donations are a very strong signal to other donors. No self-donation makes a fundraising page much less trustworthy, and a self-donation confirms the overall story that the fundraiser is telling - namely that this cause is worthwhile and deserves your donation.

Commitment

The more time and resources we commit to an effort, the higher propensity that we will escalate our commitment even further. This is the underlying principle behind the human behavior pattern of commitment escalation.

It is precisely that pattern we try to trigger by getting a fundraiser to self donate. If you self-donate, you have already committed yourself more, and therefore are more likely to put in that extra effort to get others to donate.

But maybe the effect goes the other way too and only those who already are committed will self donate? If this was the case we would see the same result as above, but we would yield nothing from encouraging more self-donations. 

To figure out if this is the case, we will look at some more advanced data analysis.

A model of fundraiser success

We recently put in the work to build a model of fundraiser success. This model is a machine learning model that helps us generate individual recommendations for P2P fundraisers.

Part of this model is a regression model over which factors impact the total fundraised amount. Here self-donation also shows itself as the biggest contributor to the final result. It is the number one feature that impacts the final result.  

In this model, we have over 30 other variables to explain the final fundraiser result. These would also capture the initial commitment of a supporter, e.g. the length of the description written, the number of images added, if they have written their description, and so forth. Thus the effect of self-donation is ‘everything else equal’ - in other words, we are quite sure that the effect isn’t caused by those self-donating already being the most committed individuals. 

But maybe other factors factor in when explaining why self donations have such a large impact? Next, we will investigate this by looking at how the fundraiser’s success depends on the size of the self-donation.

The size of the self-donation

In the below graph we can see the impact of the self-donation on the final result measured over the size of the self-donation (for various reasons it is measured in Danish kroner DKK - but the data covers both Danish, Scandinavian, and European charities). 

It isn’t possible to interpret the value of the impact, so we are only looking at the slope of the curve and if it is a positive or negative effect.

As we can see the impact of the self-donation plateaus at 600 DKK (80 €). More surprising is that the effect is slightly negative for self donations below 300 DKK (40 €).

This has made us think that the partial dependency we see here is capturing something we don't have in our regression model. Namely the wealth/income of the fundraiser and especially the wealth/income of the fundraisers' friends, family, and colleagues. 

Wealthy people tend to have wealthy friends and colleagues. When a person makes a large self-donation, it both indicates that they are wealthy and that their friends are wealthy. Reverse, if they make a low self-donation, it reveals that there is a likelihood that they are not as well off as they could be.

This is just a hypothesis at this point, but it does mean that we don't encourage additional self donations from certain fundraisers, even though our machine learning algorithm suggests this. We don't want to do this, as we fear it might stigmatise an individual who get this recommendations, but don't have the financial means to follow the recommendation.

The role of self-donations in P2P fundraising

Self donations play a crucial role. They are an incredibly important signal for other donors, that the fundraiser is serious and has put their money where their mouth is. In addition, it helps escalate the fundraisers’ commitment to fundraising. This is why self-donating fundraisers are 3,7 times as likely to reach their goal as those who don’t self-donate.

We see that the impact of a self-donating increase by the amount given. This effect might have more to do with a wealth effect where large self donations indicate that the fundraisers and their social network are quite well off. 

It doesn't by any means remove the positive effect of self-donations on the propensity of fundraisers to reach their goals. But it does indicate that part of the effect on the total amount fundraised has to do with wealth rather than self-donation as a signalling and commitment escalating tool.