Updated on August 6, 2015

The Internet is awash with guides for finding success on the crowdfunding website Kickstarter. A quick search yields (in numerical order):

  • “6 Tips From Kickstarter on How to Run a Successful Crowdfunding Campaign”

  • “Crowdfunding Secrets: 7 Tips For Kickstarter Success”
  • “8 Things I Learned From My (Failed) Kickstarter Campaign”

  • “Kicking Ass & Taking Donations: 9 Tips on Funding Your Kickstarter Project”

  • “10 Tips I Wish I Knew Before I Launched My Kickstarter Campaign”

And so on.

But the best advice to those seeking money online might sound more like this: Be thin, fair-skinned, and attractive.

It is true that in many realms, crowdfunding has delivered on its democratic promise. Take female entrepreneurship: It’s been shown that professional investors consistently view pitches from men more favorably than those from women, even when the content of those pitches was the same. Kickstarter has subverted that. On the site, projects launched by women are more likely to secure funding than those started by men.

That said, some recent research suggests that when it comes to websites that connect investors with entrepreneurs and donors with charitable recipients, the Internet’s levelling power is not strong enough to dissolve other types of longstanding bias. Fundraisers’ physical appearance, which is evident from their profile pictures on sites such as Kickstarter and Kiva, can subtly guide donation or investment decisions. That’s worrying as a trawl through beseeching online profiles has become a more and more common way for donors, investors, and shoppers to decide how to give their money not just to any given entrepreneur, but even to homeless people and prospective college students.

Recently, researchers from Hong Kong University of Science and Technology, the National University of Singapore, and Nanyang Technological University (also in Singapore) looked at how users of the microlending site Kiva decided who they’d give a loan to. (While they studied Kiva in particular, their findings are about the users of that site, and are likely relevant to how funds get distributed on other platforms that feature the profile pictures of those seeking money.) To start, the researchers coded and rated the profile pictures of thousands of potential recipients, recording their attractiveness, skin color, and physique. (These are glaringly subjective, but each picture was rated by four different people on two different continents, to make sure that ratings didn’t differ too wildly.)

After analyzing the tens of thousands of loans made in a single month, the researchers found that Kiva’s users appeared to rely on the physical characteristics visible in profile pictures when choosing a recipient. The discrimination could be quantified: For a loan of $700, the average size on Kiva, the beautiful received the equivalent of a $60 bonus. Recipients who were overweight or had relatively dark skin, on the other hand, would suffer penalties of roughly $65 and $40, respectively. (These bonuses and penalties aren’t insignificant, but the researchers uncovered predictors of funding that were even stronger. In general, fundraisers who were women or from the world’s poorest regions were the quickest to get funded.)

Even though Kiva has charitable elements, picking a recipient is still an investment: Even a purely charitable lender will try to maximize the chances of recovering his or her loan, if only to make another one on the site. Keeping that in mind, what if Kiva users are consciously seeking out the beautiful, the slim, and the fair-skinned because they think those types of people will be more likely to repay their loans on time? The researchers investigated this possibility, and found that these characteristics aren’t useful predictors of whether a recipient will pay back his or her loan. While Kiva’s users might not be consciously unfair, their biased decision making doesn’t lead to any smarter investments.

Regardless of whether such bias is conscious or unconscious, the researchers had a guess as to why some users were being swayed by physical characteristics. “We argue that implicit discrimination may characterize lending decisions on Kiva, because of the dizzying array of choices available and the lack of any obvious decision criteria for making a funding choice,” they write. Without anything else to go on, users default to old stereotypes. Indeed, it turned out that the same investment biases didn’t appear when the researchers analyzed only the loans made by the most experienced of Kiva’s lenders. And in the years since 2009, when the Kiva data was from, the site has implemented some features to reduce the feeling of choice overload, including badges on fundraisers’ profiles that indicate an emphasis on savings, entrepreneurship, or other specific values.

The perks of asking for money while beautiful don’t stop at Kiva—earlier this year, University College London’s Nichola Raihani and the University of Bristol’s Sarah Smith uncovered a similar effect in the fundraising efforts arising from the London Virgin Marathon. (Unlike in the Kiva study, Raihani and Smith focused on perceived attractiveness alone, not skin tone or body size.) As with many distance-running events, marathoners were encouraged to raise funds for various causes, and could set up pages with their fundraising goals and a list of donor names and their pledged amounts. As in the Kiva study, the researchers recorded recipients’ gender and attractiveness. Then, they closely examined the donations that accumulated on each page.

They noticed an interesting pattern. It wasn’t just that more-attractive runners of both genders brought in more money (which they did, netting nearly $300 more on average). It was that the order and size of the previous donations mattered: Whenever men donated a lot of money (“a lot” being twice as much as the average donation on that page) to an attractive woman’s cause and then were listed as the most recent donor on her page, the next man who donated tended to pledge an even larger amount of money. Ignoring gender, the typical follow-up contribution when one person donated a lot of money was about £10 larger than the average donation; when it was a man following another man’s large donation on an attractive woman’s page, the increase was closer to £30. Men gave four times as much in this case than they normally did following a large donation to a less-attractive woman. (Women demonstrated no such pattern.)

Men Donate More When Trying to Show Off on an Attractive Woman’s Profile

Raihani and Smith

To make sense of this behavior, Raihani and Smith thought of each donation page as a “tournament” in which men vied for the attention of attractive female runners. The researchers wrote of “a biological market, where individuals compete for access to partners with the highest market value by signaling their value through costly helping displays.” Whether the men knew it or not, the theory goes, they were using big donations to signal that they were wealthy or that they were cooperative—both of which are sought-after traits on the evolutionary market.

Parts of the online marketplace, then, are subject to the same old biases that reign in the analog world. Just as the Kiva study suggested, door-to-door fundraising experiments in the past have shown that attractive women and non-black people have an edge. And just as the marathon study revealed, men playing a small game in a lab were shown to be more generous when a woman was present—and even more so when that woman was attractive.

Some biases may be even stronger on the Internet. A study published earlier this year catalogued a series of interactions on World of Warcraft, and found that the appearance of players’ avatars ended up shaping how likely they were to get receive help from others in the game. Even though avatars are customizable and offer no useful indication of a player’s physical appearance (let alone personality traits), the more-attractive ones tended to get more help. And when female players picked male avatars, they received less public goodwill than male players who picked male avatars. Which suggests a corollary to an old saying: On the Internet, everyone knows you’re a woman.