Wikipedia:Primary topics with respect to usage
 | This is an essay. It contains the advice or opinions of one or more Wikipedia contributors. This page is not an encyclopedia article, nor is it one of Wikipedia's policies or guidelines, as it has not been thoroughly vetted by the community. Some essays represent widespread norms; others only represent minority viewpoints. |
Usage is one of the two determiners of a primary topic, the other being long-term significance. The guidelines define a primary topic with respect to usage as the one that is highly likely—much more likely than any other single topic, and more likely than all the other topics combined—to be the topic sought when a reader searches for that term.
There are no easy ways to reliably measure usage. However, the following may help:
- Clickstream data. This is available at meta:Research:Wikipedia clickstream. It is a dataset showing how many readers have clicked on links from one article to another. The data is not very user-friendly (accessing it requires querying a very large text file), but can be extremely useful, for example, in gauging what links on a given dab page readers follow. The data is available on a monthly basis, for source–destination pairs that have received at least 10 clicks for the month. The most recent two months of data are visualised in Wikinav.
- Pageviews. There are easily accessible tools to display the pageviews – how many times each article, or redirect, has been viewed in a given period. Suggestions for how these can be used are available at Wikipedia:Pageviews and primary topics. However, except for special cases, the usefulness of pageviews is limited. Readers searching directly for the term in question usually account for a small proportion (typically well below 10%) of the total traffic, the rest being due to incoming links. So, for example, if one of two articles gets 3x more views as the other, this does not mean that 3x as many readers will be searching for it – the ratio among the 10% of views due to direct searches will rarely match the ratio among the 100%. Pageviews may give a very rough indication, but it is only reliable if the difference in pageviews is massive.
- Piping experiments. To find out how many readers click on a given link, that link can be temporarily piped via a redirect that otherwise receives no views. The pageviews for this redirect will then show the number of people following that link.
Usage may be different from long-term significance, but it is still generally understood to be relevant primarily for the longer term. An article may become very popular for a short period of time because of prominent news coverage, but this popularity will likely be irrelevant for determining usage unless it has been sustained for a long time.