There's no doubt about it: the appeal from Jimmy Wales is a strong message. We've tested it head-to-head against other banners, and the results  are unequivocal - especially when you also compare its performance last year and the year before.
But nobody wants to just put Jimmy up on the sites and leave him up for two months!
So we're issuing a challenge: Find the banner that will beat Jimmy.
Data informed conclusions
Here's the trick:
We have to make our decisions based on the facts, not our instinct. Please read the summaries below for really important details from our focus group and survey of past donors.
Wikimedia conducted a focus group of past donors in the New York City area in September 2010. It's important to note that this was a single focus group, and in a single city. We'll need to do more to make sure that results correlate universally. But we came out of it with a few important take-away points. It's important to realize that these points reflect ONLY donors - they should not be read as a wider feeling about mission or strategic direction - they're messaging points to help us refine and deliver the best messages possible.
** The most powerful image is of Wikipedia as a global community of people who freely share their knowledge and self-police the product.
For everyone who participated, the idea of a global community of people sharing knowledge that is accessible to anyone who wants it free of charge is incredibly powerful. Respondents in this group were highly unlikely to be editors themselves; most consider themselves users. They love the idea of the community and want to support it, but they are reluctant to put themselves out there by being more than a user and a donor.
** Keeping the projects ad-free is a powerful motivator.
Respondents were unanimous that keeping Wiki[m\p]edia ad free should be a priority, even if it meant that Wiki[m\p]edia would be approaching them for money more often. Accepting paid ads could corrupt the values and discourage the free flow of information.
** Independence is critically important.
These respondents consume a lot of media, and they place a high premium on the free flow of information. They have little patience for “sponsored” news or information that excludes other perspectives. The Wikimedia model of openness and community engagement facilitates that.
** It’s a cause because it’s a tool.
This may sound a bit like a chicken/egg argument, but it’s actually an important nuance. These folks use Wikimedia every day for things from simple curiosities to serious research. So it’s a tool that lets them get what they need. But it has grown to 17 million articles in 270 languages. Because it has that kind of depth and it reaches so many people around the world, it’s worth protecting what the community so successfully built. And that makes it a cause too.
** Growing isn’t always a good thing, when positioning for donors.
Like many tech savvy folks, our respondents are a suspicious lot. The idea of Wikimedia growing brings up concerns about what Wikimedia would become, and fears about the path of companies like Facebook. It’s not just a privacy concern; it’s a concern about what would happen to the democratic model of Wikimedia inside a growth strategy. Supporting the organic growth of the community doesn’t raise the same concerns.
** Supporters strongly reject any agenda being attached to Wikimedia, even when that agenda would extend the current offerings.
An agenda implies ownership, and respondents feel pretty strongly that the community owns Wikipedia. They think of Wikipedia as an organic thing, not like a typical nonprofit, and any attempt to steer it would disrupt that. Community support is one of the key values, and not everyone in the community would support new initiatives.
** There is room to fundraise more aggressively.
Across the board, respondents were surprised that they didn’t have the opportunity to give to Wikimedia more often. Obviously, there is a balance and a PBS-style solicitation schedule wouldn’t make sense both for Wikimedia’s personality and for this audience, but there is much more space available than we are taking.
** Wikimedia donors are highly suspicious of marketing gimmicks.
Simple, direct messages are likely to work best. Jimmy’s message worked not so much because he was the founder, but because it was a simple plea for support delivered authentically.
As we know, that’s something that also needs quantitative testing to prove. Sometimes donor response in a focus group and donor activity don’t line up exactly. But, some things already line up with early tests. The more gimmicky the banner, the less likely it is to drive donations even if it increases clicks.
Reaction to banners like “572 have donated in New York today” also raised concerns about privacy – not a good reaction in an already suspicious audience. Appeals to “keep us growing” or that highlight a contributor’s work raise earlier concerns about an agenda.
Donor Survey Highlights
Wikimedia produced a random sample of 20,000 individuals from the much larger number of individuals, from many countries, contributing less than $1000 between November 1 2009 and June 30 2010. These individuals were invited to participate in a 29 item (but around 70 question) survey. 3760 agreed to participate, and the survey was conducted in August 2010. The participants probably differ from those who declined in ways that are associated with survey answers. Hence the respondents do not represent an entirely representative sample of the < $1000 donors.
The survey participants are committed to Wiki[p/m]edia, visiting it frequently. They say that they are very likely to donate again, and they support all the survey-mentioned reasons for donation. They were not aware of Wikipedia chapters. A majority of respondents did not appear greatly concerned about possible threats to Wikipedia’s identity.
About 1/3 of these individuals have edited, though not frequently. Those who express more support for Wikimedia as a cause appear more prone to edit. Those who have not contributed in this way say mostly that they haven’t thought about it--suggesting that they haven’t really considered the possibility—or that they don’t have time. Europeans and the highly educated especially stress lack of time.
Some subgroup differences were found within the sample. The likelihood of writing or editing does vary a bit by subgroup, for example. Overall, however, responses did not vary greatly by subgroup, whether “demographic” (nationality, education, sex) or behavioral (e.g., degree of on-line activity).
Chapters will receive the specifics of how we will work with them through their fundraising contacts which were designated on the fundraising survey, in order to keep the information communicated here to the essentials.
We have been testing for ten weeks now, and are really pleased with the progress that the tech team has made with new tools to support the fundraiser. Geotargetting appears to work now, and we are currently testing a 1 step versus 2 step donation process. We will have solid test results this week, we believe. In all, we believe that we are - technically and message-wise - in a really good position. We're working out kinks, definitely, but we're working them out before the fundraiser starts, so that we can maximize the dollar-earning potential of every day that we have banners up.
Thanks for sticking through this email - join us in discussion and help us beat the Jimmy appeal!