I have added forecasting to my report but the results do not make sense. What is happening here?
Forecasting is a result manipulation that uses patterns in your data to predict future results. For information on adding forecasting and more information on how the forecasting works see using forecasting to predict results. At its most basic, Explore forecasting is a two steps process. First it looks for a pattern that describes your historic data, second it applies that pattern forward in time. This basic approach allows forecasting to be applied to a wide range of data but it also has limitations.
Developing an effective forecasting model requires extensive testing, understanding of the data being modeled and development. Explore forecasting is not meant to be a fully developed forecasting system.
Explore forecasting works best when the data is composed of a trend and a cycle on top of the trend. For example, if your ticket volume is growing over time but you also have surges in ticket volume around holidays forecasting may do a good job of predicting ticket volume during next year's busy period.
Data which is a function of multiple inputs is not a good candidate for forecasting. For example, first reply time is a function of ticket creation, staffing, agent efficiency etc. These variables are not captured by Explore forecasting and as a result the forecast will likely not be useful.