diff --git a/R/adjust-equivocal-zone.R b/R/adjust-equivocal-zone.R
index cfbc9a7..829fade 100644
--- a/R/adjust-equivocal-zone.R
+++ b/R/adjust-equivocal-zone.R
@@ -16,6 +16,15 @@
 #' this adjustment just collects metadata on the supplied column names and does
 #' not risk data leakage.
 #'
+#' @details
+#' This function transforms the class prediction column `estimate` to have type
+#' `class_pred` from [probably::class_pred()]. You can loosely think of this
+#' column type as a factor, except there's a possible entry `[EQ]` that is
+#' _not_ a level and will be excluded from performance metric calculations.
+#' As a result, the output column has the same number of levels as the input,
+#' except now has a possible entry `[EQ]` that tidymodels funcitons know to
+#' exclude from further analyses.
+#'
 #' @examplesIf rlang::is_installed(c("probably", "modeldata"))
 #' library(dplyr)
 #' library(modeldata)
diff --git a/man/adjust_equivocal_zone.Rd b/man/adjust_equivocal_zone.Rd
index 2c970ce..42f5476 100644
--- a/man/adjust_equivocal_zone.Rd
+++ b/man/adjust_equivocal_zone.Rd
@@ -20,6 +20,15 @@ too uncertain or ambiguous to be assigned a hard class. Rather than
 predicting a hard class when the probability is very close to a threshold,
 tailors using this adjustment predict "[EQ]".
 }
+\details{
+This function transforms the class prediction column \code{estimate} to have type
+\code{class_pred} from \code{\link[probably:class_pred]{probably::class_pred()}}. You can loosely think of this
+column type as a factor, except there's a possible entry \verb{[EQ]} that is
+\emph{not} a level and will be excluded from performance metric calculations.
+As a result, the output column has the same number of levels as the input,
+except now has a possible entry \verb{[EQ]} that tidymodels funcitons know to
+exclude from further analyses.
+}
 \section{Data Usage}{
 
 This adjustment doesn't require estimation and, as such, the same data that's