%md
There’s not much context because we don’t have data from other days, but for the sake of this exercise, let’s assume that the number of clusters more than tripled any other day and the number of users did not change meaningfully during that time period. If this were truly the case, then the only reasonable explanation would be that the clusters were likely created programmatically using jobs. Additionally, 12/28/19 was a Saturday, so we don't expect there to be many interactive clusters created.
Since we know that the tag in the field tell us who / what created a cluster, we can use to see what caused the spike.
CLUSTERS
Last refresh: Never