Value of a Wikipedian

Every organization which recruits volunteers values their volunteers. Sometimes they even calculate the financial value of their volunteers. This works by recording the cost of a set of volunteer recruitment campaigns, noting how many volunteers become engaged because of those campaigns, then dividing the total cost by the number of volunteers recruited.

I am sure that the Wikimedia Foundation must know this number but so far as I know it has never been explicitly stated. When outcomes of projects are reported, most commonly by Wikimedia chapters or other grant recipients who are required to do so as a result of getting funding from the Wikimedia Foundation. Funds are dispersed through various channels. Funds to chapters go with approval from the Wikimedia Foundation board and typically at the behest of the Funds Dissemination Committee (FDC), which is a community-elected group of Wikimedia volunteers who review the proposed budgets of Wikimedia chapters and offer them funding based on a review which is supposed to have its basis in traditional nonprofit management and established conservative nonprofit grantmaking practices. Smaller amounts of funding, typically to individuals or small groups, is made by staff at the Wikimedia Foundation who again try to practice conservative grantmaking but who can make quicker decisions than the FDC because they are knowledgeable professionals doing this full time with smaller money amounts.

Measuring impact is something that every organization wishes to do but which all organizations find difficult. In the Wikimedia community, the desired impact of grantmaking is

• more individuals
• representing more demographics
• contribute more Wikimedia content
• in more languages
• and of more content types
• with more regularity and frequency
• and of higher quality
• while interacting with more other Wikimedia contributors

I am becoming more interested in learning how much is invested to recruit and develop volunteer Wikipedians, and how that money amount would compare to hiring people outright. The budget of the Wikimedia Foundation has been as follows:

• 2001 established – no budget
• 2005 1.5 million USD, hired first employee
• 2008 8 million
• 2009 18 million
• grows more…
• 2013 53 million
• 2015 – probably around 65-70 million

Most of the community discussion about not having money for paid staff happened between 2005-08, and has not been re-examined in the context of the Wikimedia Foundation having much more money now and the growing demand for more content of higher quality being presented more quickly. There is a lot of talk about stagnation in community growth. If it could happen that a few paid staff could do the sorts of things which would support volunteers, then I hope that those kinds of things could be funded. Last year in the context of Wikipedians in Seattle having some paid staff, I listed some reasons why I thought this would be useful. My argument was that certain office actions – particularly collecting metrics for the Wikimedia Foundation and responding to Wikimedia complaints unrelated to one’s own actions – are not things that volunteers enjoy doing but things which the Wikimedia Foundation pressures volunteers to do. The pressure from the Wikimedia Foundation that volunteers do certain unpleasant things is a deterrent from volunteers engaging more deeply, and since at similar nonprofit organizations these kinds of things are typically done by paid staff, and since the money is available, and since more impact is desired, and since paying staff is probably more cost effective than trying to use the same funds to recruit a volunteer to do the unpleasant work for free, then why not pay the staff? I still support the Wikimedia Foundation not paying staff to produce content, even though I wish that there were more infrastructure in place for other interests to fund the creation of content for Wikimedia projects.

I am going to guess a number based on my experience, and perhaps this number can be corrected someday. I think that if money is invested to recruit and develop a Wikipedian, then USD 10,000 is what must be invested to produce a good Wikimedia contributor who engages with projects 4 hours weekly for a year. I expect that more likely this dollar amount is low than high. I think that this dollar amount is international – although costs are higher in countries like the United States, people are more likely to be willing and ready to volunteer, even though this amount is about 1/5 of a lower middle class annual salary. In a country like India where USD 10,000 is a full annual middle class salary, the outreach costs are still the same, because for various reasons especially including that Wikimedia projects seem foreign and are not integrated with cultural expectations, the costs of impact from outreach are higher, and still I think it costs about the same to recruit a similarly engaged volunteer.

There are many variables in determining the value of Wikimedia volunteers and the development of infrastructure to serve a particular demographic can greatly decrease the cost per volunteer in recruitment campaigns. People with metrics who are looking more closely at these things can have better metrics than me. The most effective kind of Wikimedia outreach remains passive outreach – the site is set up, and many people go to it to read, and dependably some percentage of those will continue to try to engage more without anyone developing personal outreach campaigns to target them, or without having anyone geographically close to them as a point of local contact.

If anyone wanted to do their own research about the impact of funding, this is easier to do for projects funded by the Wikimedia Foundation than it is for any other nonprofit on this scale and at this grassroots level. I believe this because I think that the documentation of projects funded by the Wikimedia Foundation is deeper and easier for the public access than for any other similar funding initiative that I have seen, and I feel like I have seen enough funding and reporting models for grassroots activism to speak authoritatively on this topic.

There are a lot implications of what I am saying, but I want to draw just one conclusion for now. I often talk and think about South Asia, particularly India, Bangladesh, Nepal, and Pakistan. Right now, the Wikimedia Foundation is willing to fund projects to support volunteers but not fund staff to do work outright. One of the disconnnects between the Wikimedia Foundation’s policy and what I think is most practical is that often, they are willing to fund projects to attempt to recruit volunteers even when the expected outcome is likely to be more expensive and have less impact than hiring someone outright. In the case of South Asia, for example, they are unwilling to hire staff community organizers to be managed by the local community or volunteer chapter. They are even more unwilling to hire staff to contribute content to Wikimedia projects, like for example the digitization of museum or archival collections or other seed content for underserved languages which some people feel would set a good precedent and example for volunteers to emulate. They are willing to fund attempts to recruit volunteers who would do these kinds of things. The hope, although not explicitly articulated, is that the outreach efforts will identify volunteers who will do more than what paid staff would accomplish. In my opinion, the reality is that outreach is expensive and considering the economy in South Asia and the local culture about volunteering, more good would come from paying people outright to contribute content. For some languages, a staffperson could be hired for 1-2 years at a rate of USD 10,000/year, which is a respectable salary in some library fields for someone at the master’s degree level. In the United States a comparable person would cost USD 70,000 at least. Additionally, for some Indian languages, contributing existing media could increase the size of some Wikipedias by 5% – a huge jump in available content which would never be possible in the short term by one person for any developed Wikipedia project like English or other European languages. The need is also greater in South Asia, as the population is so high. Finally, necessary gaps in representation could be closed by hiring paid staff. There is a concern, for example, that women are underrepresented in Wikimedia projects. This is true, but I regret that actual numbers are not often articulated because the problem is discussed in a silly way. It would be closer to the truth to say there are not many regular Wikipedia contributors of any gender. To take one of the most developed South Asian languages as an example, Bangla, if the Wikimedia Foundation were to fund 20 women at USD 5000 per year to contribute to Wikimedia projects, then they would be creating material in the sixth-month spoken language in the world. Honestly, there are not more than 20 highly active contributors in Bangla language. Many people contribute a little – but people who are available to help with administration are much fewer. If only 20 women with library skills were paid to regularly contribute and join the conversation, and if they actually participated in a modest way of the sort that casual male volunteers would do, then the intervention really could change the direction of Bangla language media and the nature of Wikimedia development. USD 5000 a year is enough to hire respectable contributors in East Bengal and Bangladesh (potentially full time for some positions, or part time for the most skilled people), if only that were an option which could even be considered. I have a lot of respect for the outreach projects which happen instead, but if the goal were to achieve impact rather than adhere to the ideology that the Wikimedia Foundation must never hire contributors – then I think that hiring some contributors in some cases would be the obvious path the movement development. I will say again – I think it costs USD 10,000 to recruit a good volunteer in the United States, and the same amount to recruit one in Kolkata, and some money should be spent to hire staff outright.

Incidentally – the Wikimedia Foundation hires consultants which I think bill at USD 125 an hour to do community tasks that they think are too sensitive for volunteers to manage. I do not want to identify anyone because they do good work but still it makes me uneasy to think that hidden professionals can be paid but anyone with a volunteer affiliation cannot get a stipend or part-time funding. I would even favor setting up an outsourcing communications and content creation office in India, because I think that is a viable model for producing baseline content and ensuring that Wikipedias there have minimal viability as reference works.