Wikipedia:Content assessment

The following system is used to assess the quality of a Wikipedia article. The system is based on a letter scheme that reflects principally how factually complete the article is, though language quality and layout are also factors.

The quality assessments are mainly performed by Wikipedia editors, who tag Talk pages of articles; some bots set the class tag for assessment based on other Talk tags or based on editor selected values. Editor assistance tools like Rater apply automated ORES or Lift Wing article assessments, offering a prediction based on structural characteristics of the page (e.g. sectioning and references) that correlate with quality, for the class tag. These tags are then collected by a bot, which generates output such as a log and statistics. For more information, see Using the bot. In 2023, project-independent quality assessments were introduced, so editors only have to rate an article once and it applies to all associated projects.

Most grades are assessed by individual editors according to the criteria on this page. Generally speaking, all editors, including editors who have written or improved an article, are encouraged to boldly set any quality rating that they believe is appropriate, except for the GA, FA, and A-class ratings. GAs (Good Articles) are generally reviewed by a single independent editor after a nomination at WP:Good article nominations. FAs (Featured Articles) are reviewed by several editors at WP:Featured article candidates. Be aware that a few projects have opted out of the standard quality scale, and use their own variation of the criteria more tuned for the subject area, such as Wikipedia:WikiProject Military history/Assessment.

It is vital that editors not take these assessments of their contributions personally. It is understood that we each have our own opinions of the priorities of the objective criteria for a perfect article. If there is disagreement over the quality rating of an article, then it should be discussed on the article's talk page.

As of August 2024, over eight million articles have been assessed. Several other languages are also using this assessment system or a derivative thereof.