Summary: Small wikis have a challenge of attracting readers and editors. Most visitors to wiki sites come from search engines. So, existing content plays a big role in attracting new visitors and editors. This is a seemingly chicken-and-egg problem.
In this situation, how do we maximise new editors for a given amount of effort/spending?
The idea is:
a) to bootstrap projects like Wiktionary, Commons, Wikisource etc., with content acquisition
b) to treat all the sister wiki projects in a language as one portal
c) do search engine optimisation for projects that lend well
d) funnel visitors and editors from "peripheral" wikis to editor-intensive ones.
- Wiki projects need content to attract editors.
- Content for some projects like Wiktionary, Wikisource, Commons etc., can be acquired by way of lobbying governments, museums, etc. Therefore, it is easier to bootstrap such projects even when the community is small.
- With proper Search Engine Optimisation, users will come to these sites from search engines.
- We then cross-link pages between the various projects to funnel visitors to editorial-intensive projects like Wikipedia
My experience with Tamil Wiki projects was a clear motivation (although the method outlined is applicable for many smaller wikis):
- We noticed that a significant number of users come to Wikipedia from search engines. In fact, a bulk of them come from porn-related queries. We even wrote quality encyclopedic articles not wanting to miss those users. The article on Orgasm (started by me) is the top-17 most visited article on Tamil Wikipedia.
- We were skeptical when ta:User:Ravidreams wanted to spend some time bootstrapping Wiktionary even when there were very few editors in Tamil Wikipedia. (But, it has paid off later, because, Wiktionary doesn't cannibalise Wikipedia editors, but attracts people who would have otherwise not edited any other wiki.)
- I wrote a bot (with Ravi's help) to get public domain dictionaries and uploaded 100,000 entries that showed the content-acquisition potential for these projects.
- We're yet to do funneling using templates.
- Have we tested the usefulness of cross-linking as a funneling mechanism?
- No. We need to do some A|B testing for that. However, articles seeded by en:User:Rambot and en:User:Ganeshbot have shown that SEO due to acquired content does attract readers and editors. See en:Special:RecentChangesLinked/User:Ganeshbot/Created for example. I can see that, in Marathi Wikipedia, they have created one word stub entries in an apparent effort to attract search engine-referred users. It seems the effort has even paid off to a certain extent. However, if this can be realised without compromising on Wikipedia article quality using the methods that I have outlined, it would be much better.
- Toolserver requirement
- "Web Analytics & Visitor Behavior Analysis". Spartan Internet Consulting. http://www.spartaninternet.com/web_analytics.asp. Retrieved 2010-09-26.
- "Wikipedia article traffic statistics". stats.grok.se. w:en:User:Henrik. http://stats.grok.se/ta/201008/%E0%AE%AA%E0%AF%81%E0%AE%A3%E0%AE%B0%E0%AF%8D%E0%AE%9A%E0%AF%8D%E0%AE%9A%E0%AE%BF%E0%AE%AA%E0%AF%8D%20%E0%AE%AA%E0%AE%B0%E0%AE%B5%E0%AE%9A%E0%AE%A8%E0%AE%BF%E0%AE%B2%E0%AF%88. Retrieved 2010-09-26. "w:ta:புணர்ச்சிப் பரவசநிலை/புணர்ச்சிப் பரவசநிலை has been viewed 1176 times in 201008. This article ranked 17 in traffic on ta.wikipedia.org."
- ta:பயனர்:Sundar/விக்சனரி_தானியங்கித்திட்டம் (in Tamil)
Do you have a thought about this proposal? A suggestion? Discuss this proposal by going to Proposal talk:SEO Market and Contributor Funnel.