diff --git a/DESCRIPTION b/DESCRIPTION
index d18429e..2e5e4a9 100644
--- a/DESCRIPTION
+++ b/DESCRIPTION
@@ -1,6 +1,6 @@
 Package: efun
 Title: Miscellaneous Functions by E
-Version: 0.3.2
+Version: 0.3.3
 Authors@R: 
     person(given = "Eduardo",
            family = "Alfonso-Sierra",
diff --git a/R/assert.R b/R/assert.R
index c26fac4..d33c683 100644
--- a/R/assert.R
+++ b/R/assert.R
@@ -27,36 +27,32 @@
 #'   is thrown.
 #'
 #' @export
-assert <- function(.data, ..., msg = "Assertion does not hold") {
-  # TODO: change this old approach and embrace more dplyr by something like
-  # abcd <- dplyr::mutate(.data, ...)
-  # and then just extract the evaluated conditions to report. This also
-  # allows multiple conditions, unlike the original approach
-  condition_eval <- with(.data, ...)
-  if (all(condition_eval)) {
+assert <- function(.data, ..., msg = "Assertion does not hold", na.rm = TRUE) {
+  conds <- rlang::enquos(..., .named = TRUE)
+  conds_eval <- purrr::imap(conds, \(x, idx) rlang::eval_tidy(x, data = .data))
+
+  if (all(purrr::map_lgl(conds_eval, \(x) all(x, na.rm = TRUE)))) {
     return(.data)
   } else {
-    msg <- paste0(
-      msg, ": `", deparse(substitute(...)), "` is false ",
-      sum(!condition_eval),
-      " out of ", length(condition_eval), " times (",
-      sprintf("%.1f%%", 100 * sum(!condition_eval) / length(condition_eval)),
-      ")\n"
-    )
-    cat(msg)
+    fail_msg <- purrr::imap(conds_eval, \(x, idx) {
+      n_fail <- sum(!x)
+      if (n_fail > 0) {
+        paste0(
+          "\t`", idx, "` is false ",
+          n_fail, " out of ", length(x), " times (",
+          sprintf("%.1f%%", 100 * n_fail / length(x)), ")\n"
+        )
+      }
+    }) |> purrr::compact() # we wanto to get rid of NULL, als error msg
 
-    NULL -> .condition_eval
-    .assert_fails <- dplyr::mutate(
-      .data |> dplyr::ungroup(),
-      .condition_eval = condition_eval
-    )
+    .assert_fails <- dplyr::mutate(.data, ...)
     if (rlang::is_interactive()) {
       utils::View(.assert_fails) # TODO: should we sample?
     } else {
       .assert_fails |>
-        dplyr::filter(!.condition_eval) |>
+        # dplyr::filter(!.condition_eval) |> # TODO: should we keep filtering?
         print() # TODO: should we sample?, or just rely on tibble's print?
     }
-    stop(msg)
+    stop(msg, "\n", fail_msg)
   }
 }
diff --git a/tests/testthat/test-assert.R b/tests/testthat/test-assert.R
index a720448..1b97f57 100644
--- a/tests/testthat/test-assert.R
+++ b/tests/testthat/test-assert.R
@@ -1,12 +1,42 @@
-test_that("assert works", {
+test_that("assert() works", {
   expect_error(mtcars |> assert(mpg < 33))
 
   expect_identical(mtcars |> assert(mpg < 34), mtcars)
 })
 
-test_that("assert does not fail on grouped data frames", {
+test_that("assert() does not fail on grouped data frames", {
   expect_error(
     object = mtcars |> dplyr::group_by(cyl) |> assert(hp < 150),
     regexp = "Assertion does not hold"
   )
 })
+
+test_that("assert() does not fail testing data with NAs", {
+  expect_no_error(
+    object = mtcars |>
+      dplyr::mutate(mpg = dplyr::na_if(mpg, 15)) |>
+      assert(mpg > 10)
+  )
+})
+
+test_that("assert() can handle multiple conditions", {
+  expect_no_error(
+    object = mtcars |>
+      dplyr::mutate(mpg = dplyr::na_if(mpg, 15)) |>
+      assert(hp < 1500, am < 100)
+  )
+
+  expect_error(
+    object = mtcars |>
+      dplyr::mutate(mpg = dplyr::na_if(mpg, 15)) |>
+      assert(hp < 1500, am < 1),
+    regexp = "Assertion does not hold.*am < 1.*"
+  )
+
+  expect_error(
+    object = mtcars |>
+      dplyr::mutate(mpg = dplyr::na_if(mpg, 15)) |>
+      assert(hp < 1, am < 100),
+    regexp = "Assertion does not hold.*hp < 1.*"
+  )
+})