diff --git a/NAMESPACE b/NAMESPACE
index 91e8700fe..94a224678 100644
--- a/NAMESPACE
+++ b/NAMESPACE
@@ -10,6 +10,7 @@ S3method(as_tibble,dtplyr_step)
 S3method(auto_copy,dtplyr_step)
 S3method(collect,dtplyr_step)
 S3method(compute,dtplyr_step)
+S3method(count,data.table)
 S3method(count,dtplyr_step)
 S3method(dim,dtplyr_step)
 S3method(dim,dtplyr_step_first)
@@ -79,6 +80,8 @@ S3method(slice_tail,dtplyr_step)
 S3method(summarise,data.table)
 S3method(summarise,dtplyr_step)
 S3method(tail,dtplyr_step)
+S3method(tally,data.table)
+S3method(tally,dtplyr_step)
 S3method(tbl_vars,dtplyr_step)
 S3method(tbl_vars,foo)
 S3method(transmute,data.table)
@@ -135,6 +138,7 @@ importFrom(dplyr,slice_min)
 importFrom(dplyr,slice_sample)
 importFrom(dplyr,slice_tail)
 importFrom(dplyr,summarise)
+importFrom(dplyr,tally)
 importFrom(dplyr,tbl_vars)
 importFrom(dplyr,transmute)
 importFrom(dplyr,ungroup)
diff --git a/NEWS.md b/NEWS.md
index 528acc7b8..6a39b5ca6 100644
--- a/NEWS.md
+++ b/NEWS.md
@@ -1,5 +1,9 @@
 # dtplyr (development version)
 
+* Loading `dtplyr` doesn't break the `count()` method on data.tables anymore (@mgirlich, #201).
+
+* `tally()` is now translated (@mgirlich, #201).
+
 * More tidyr verbs are in the process of being added:
 
   * `drop_na()` (@markfairbanks, #194)
diff --git a/R/count.R b/R/count.R
index 46f86730c..ed730ce97 100644
--- a/R/count.R
+++ b/R/count.R
@@ -21,8 +21,20 @@ count.dtplyr_step <- function(.data, ..., wt = NULL, sort = FALSE, name = NULL)
     out <- .data
   }
 
-  wt <- enexpr(wt)
-  if (is.null(wt)) {
+  tally(out, wt = !!enquo(wt), sort = sort, name = name)
+}
+
+#' @export
+count.data.table <- function(.data, ...) {
+  .data <- lazy_dt(.data)
+  count(.data, ...)
+}
+
+#' @importFrom dplyr tally
+#' @export
+tally.dtplyr_step <- function(.data, wt = NULL, sort = FALSE, name = NULL) {
+  wt <- enquo(wt)
+  if (quo_is_null(wt)) {
     n <- expr(n())
   } else {
     n <- expr(sum(!!wt, na.rm = TRUE))
@@ -34,7 +46,7 @@ count.dtplyr_step <- function(.data, ..., wt = NULL, sort = FALSE, name = NULL)
     abort("`name` must be a string")
   }
 
-  out <- summarise(out, !!name := !!n)
+  out <- summarise(.data, !!name := !!n)
 
   if (sort) {
     out <- arrange(out, desc(!!sym(name)))
@@ -42,3 +54,9 @@ count.dtplyr_step <- function(.data, ..., wt = NULL, sort = FALSE, name = NULL)
 
   out
 }
+
+#' @export
+tally.data.table <- function(.data, ...) {
+  .data <- lazy_dt(.data)
+  tally(.data, ...)
+}
diff --git a/tests/testthat/test-count.R b/tests/testthat/test-count.R
index f5bd6a342..e88aa4f2d 100644
--- a/tests/testthat/test-count.R
+++ b/tests/testthat/test-count.R
@@ -41,3 +41,11 @@ test_that("can sort", {
     tibble(x = c(2, 1), n = c(10, 3))
   )
 })
+
+test_that("tally works", {
+  dt <- lazy_dt(data.table(x = c(1, 1, 1, 2)), "DT")
+  expect_equal(
+    dt %>% group_by(x) %>% tally() %>% collect(),
+    tibble(x = c(1, 2), n = c(3, 1))
+  )
+})