Article Score Varies Between Salesforce Reports and the API

There are two main score calculation methods used by Salesforce Knowledge, depending on how you access Article Scores.

Half-Life Calculation (Used in UI Reports)

When accessed via the UI (reports), Article Scores are calculated as an average view rating. Scores take into account a half-life calculation. Every 15 days, if an article has not been viewed, its average rating moves up or down. This calculation ensures that over time, older or outdated articles do not maintain artificially high or low ratings compared to newer, more frequently viewed articles.
The half-life calculation is based on current time and the last reference time of the articles. As the last reference time for an article increases, the score decreases. More recent views earn a higher score. Because scores are not run through a normalization calculation when viewed in a report, their values may exceed 100 if the article has a high frequency of recent views.

Normalization Calculation (Used in the API)

When accessed via the API, Article Scores are calculated as a weighted view. The article with the most views receives a score of 100. All other article scores are calculated relative to this highest view score.
Example: If the best-read article has 2,000 views and another has 1,000, the best-read article receives a score of 100 and the other receives 50.
The normalization calculation is applied when querying via the KnowledgeArticleViewStat object's NormalizedScore field via the API. This calculation is not applied when viewing scores in UI reports.

Why Scores Differ

The half-life calculation is applied in both situations — whether the score is accessed via a UI report or via the API. The reason scores differ is that the normalization calculation is only applied when querying via the API, not in UI reports.