Wikipedia – Free Knowledge and a 19-Year-Old Online Platform
Wikipedia is one of the original online platforms with many structural challenges, yet they are making it work
Wikipedia – the site where most research projects start, and the one no one sites as their final source – is a massive two-sided platform that has many challenges, but isn’t going anywhere anytime soon.
The user side of the platform is made up of billions of people all over the world, while the editors, or contributors, who have made more than 10 edits since Wikipedia’s founding, are a much smaller community of ~2 million people.1
Value Creation for the User
Wikimedia’s mission is to “empower and engage people around the world to collect and develop educational content under a free license or in the public domain, and to disseminate it effectively and globally.”2 And, in many ways they accomplish this, creating a lot of value for the world.
As every fourth grader and every second year MBA student knows, a Wikipedia search can really help you get going no matter what the assignment is. With a content library of 6.0 million English articles3, in just a few seconds, users doing a quick fact check at dinner or starting an in-depth research project can find up-to-date answers to nearly every basic question they may be asking.
As we all know, Wikipedia’s value has a limit though. For instance, I’m pretty sure this will be the only assignment I have ever and will ever submit with a Wikipedia article sited – more to come the limitations of Wikipedia’s value creation.
Value Creation for the Contributor
I can not fathom spending my free time editing Wikipedia pages, so I’ll just parrot the results of the editor survey done by Wikipedia showing why contributors dedicate their time to editing Wikipedia pages for the benefit of everyone else4:
- 71% like the idea of volunteering to share knowledge
- 69% believe information should be freely available
- 63% enjoy contributing to subject matters with expertise
- 60% think it’s fun
While I can’t get on board, nearly 1.8 edits are made every second, so who am I to say there’s no value created for the contributors!5
In terms of value capture, Wikipedia does not believe in advertisements on their website, so the Wikimedia Foundation, the non-profit organization that manages Wikipedia, is funded by donations.
We’ve all seen some version of this plea for donations or popped open a Wikipedia page confirming the capital city of Kansas is in fact Topeka and been asked for a small donation to keep Wikipedia running.
Ever wonder how those campaigns perform? In 2016, Wikipedia raised $82 million from donations and other contributions. This is enough to cover expenses and keep the site up and running for years to come, and given the non-profit nature of the company, that’s the extent of the value capture of Wikipedia – just keeping the lights on. Could this by why the donation campaigns actually work?
While one could argue Wikipedia is already at scale, their interest in continuing to scale scaling comes in two forms: they encourage more articles and more articles in more languages.
Wikipedia has strong cross-side network effects as users look for content, they benefit from contributors adding pages and keeping the current pages up-to-date.
Wikipedia’s growth rate is slowing substantially year over year (3.75% in 2019 vs 24% in 2009 6), as are the number of editors, but as new important topics come up, the content is still being added. For example, the Wikipedia page called “2019-20 Coronavirus Pandemic” already has over 800 citations and is approximately 100 printed pages.7
In terms of scaling to more languages, bots have automated translations and are helping expand Wikipedia’s reach around the globe.8
Given these factors, it’s safe to say Wikipedia’s network effects are still strong, even at their massive scale and nearly 20 years in to their journey.
Risk of Disintermediation and Multi-homing
Accuracy and perceived accuracy, completeness, biased content, and vandalism are all factors that make Wikipedia a less than perfect source of knowledge for the user, however Wikipedia is likely here to stay. As mentioned above, users don’t trust Wikipedia as a final source of knowledge for any complex content even though some estimates say over 99% of Wikipedia’s content is accurate. Completeness, however, is rated much lower at 69-91%.9 Given anyone can contribute to Wikipedia articles whenever they want, the content can be biased and is at risk of vandalism, requiring a strong network of contributors to catch and correct these malicious edits. Each of these factors leads both me and the 4th graders to verify content through other sources before submitting final versions of our assignments, but we still go there first. And that is the value that Wikipedia brings to the users. Contributors who fall into the “I enjoy contributing to subject matters with expertise” likely search for other, more reputable outlets to share their knowledge, but that doesn’t stop them from also editing in Wikipedia. Wikipedia assumes multi-homing for both its users and contributors and that works for them!
1 Number of Wikipedia contributors from 2001 – 2017, https://www.statista.com/statistics/265837/number-of-wikipedia-contributors-worldwide/, accessed 3/23
2 Wikimedia Foundation Mission, https://wikimediafoundation.org/about/mission/, accessed 3/23/20
3 Wikipedia: Size of Wikipedia, https://en.wikipedia.org/wiki/Wikipedia:Size_of_Wikipedia, accessed 3/23/20
4 Editor Survey, https://meta.wikimedia.org/wiki/Editor_Survey_2011/Editing_Activities, accessed 3/22/20
5 Wikipedia Statistics, https://en.wikipedia.org/wiki/Wikipedia:Statistics, accessed 3/23/20
6 Size of Wikipedia, https://en.wikipedia.org/wiki/Wikipedia:Size_of_Wikipedia, accessed 3/23/20
7 2019-20 Coronavirus Pandemic, https://en.wikipedia.org/wiki/2019%E2%80%9320_coronavirus_pandemic, accessed 3/23/20
8 Could Bots solve all of Wikipedia’s problems? https://medium.com/enrique-dans/could-an-army-of-bots-solve-all-wikipedias-problems-40703ba9d457, accessed 3/22/20
9 Study shows Wikipedia Accuracy is 99.5%, https://www.zmescience.com/science/study-wikipedia-25092014/, accessed 3/23/20
Student comments on Wikipedia – Free Knowledge and a 19-Year-Old Online Platform
Love Wikipedia and I donated when they asked to support their engineers because like you, I am not one of the people that provides content.
Since Wikipedia is like the place where most of the people start their research I wonder if it can start charging people for example just $1 a year would be enough since there is billions of people consuming info. I personally would pay for it. But if people are not willing to pay its scalability is at a big risk. Nowadays we are used to get most products for free (letting them use our data) so there is a chance that people don’t want to pay for Wikipedia.
A few other business models that they can use is advertising or sell the data. Advertising I think won’t bug people enough to make them stop using the site since we are all used to adds. Selling the data will have more serious implications.
I really believe that with the right for profit business model and additional rules for editing Wikipedia can become more than just the first place people search for info. I think that they can change the brand from “info for everything” to ” reliable info for everything”.
Thanks for making me second guess my entire article which used several Wikipedia references! The Wikipedia model is fascinating to me. If there were gross inaccuracies or bias in the articles, I am quite sure I would have picked on them at least once or twice. However, I can’t remember the last time I encountered this issue with Wikipedia. The content seems very informative, unbiased and accurate. I point this out because this could very well be the result of an interesting positive feedback loop with crowdsourcing. With more eyes on the articles, there is more chance that content will be flushed for accuracy, which in turn makes Wikipedia a more trusted source and therefore encouraging more users. This sounds great in principle, but I would have never predicted it to work just given human nature. It appears it does though. I’d be interested to know if they have any other tools that help curate the content outside of direct human interaction.