Page views

In November 2006, Wikia took a sample of page views to measure how many were from non-logged in users. 1% of 3.5 million page views were sampled. Note that this was a very small scale test and might not be representative for individual wikis.

Some findings

 * The results indicated that 90% of page views were from users who were not logged in. This figure matches Jakob Nielsen's statement that 90% of users are "lurkers".
 * For the top 30 wikis which have at least 97 samples, the % of reigstered users increases to 13.36%.
 * For the most visited wikis, the % of registered users drops - to 6.66% for the top 10 wikis or 6.80% for the top 5.
 * Some wikis have very high percentages of registered users. For the less popular wikis, this may be due simply to one or two users accounting for most of the hits during the sampling period. Looking just at those with at least 100 samples, some exceptions to 90-10 rule are:
 * High:
 * Star Wars Fanon with 70.3% registered
 * Central Wikia with 38% registered (www.wikia.com) or 48% for wikia.com.
 * Cybernations with 30.6% registered
 * Low:
 * Dofus with 3.2% registered
 * EVCHK with 3.1% registered
 * Smallville with 2.6% registered
 * 24 with 1.8% registered
 * Spanking Art with 1% registered

Jakob Nielsen quotes a similar statistic about the biggest and most visited wiki on earth, Wikipedia. On Wikipedia, more than 99% of visitors are just reading and less than 1% is contributing. So a number of 1% isn't necessarily something to worry about, it's just the normal participation inequality that you can always expect. An unusually high percentage of logged-in users (30% or more), on the other hand, may indicate that the wiki has very little readers from the outside world.

Increasing your number of active users
If your wiki has a high percentage of unregistered readers, think about how you can draw more of these people into your community and turn them from passive viewers to active editors who will help to improve your content. Nielsen suggests
 * making it easier to contribute,
 * making participation a side effect,
 * letting users build their contributions by modifying existing,
 * templates rather than creating complete entities,
 * rewarding participants for contributing, and
 * promoting quality contributors.

Stats
Note that for less visited wikis, the sample size is less than 20 and highly unreliable. The stats are more representative for those with at least 100 samples. The approximate sample sizes are shown below.


 * Note that some wikis are shown twice due to appearing both with and without the www (wikia.com, memory-alpha, uncyclopedia, wikiality, etc)