diff --git a/_questions/machine-learning-zoomcamp/misc/042_050caa0da5_avoid-keyerror-converted-rerun-delete-target.md b/_questions/machine-learning-zoomcamp/misc/042_050caa0da5_avoid-keyerror-converted-rerun-delete-target.md
new file mode 100644
index 00000000..abd849c6
--- /dev/null
+++ b/_questions/machine-learning-zoomcamp/misc/042_050caa0da5_avoid-keyerror-converted-rerun-delete-target.md
@@ -0,0 +1,19 @@
+---
+id: 050caa0da5
+question: 'How do I avoid `KeyError: ''converted''` when re-running a cell that deletes
+  the target column in pandas?'
+sort_order: 42
+---
+
+This happens because once you run a cell that deletes `converted` (e.g., `del df_train['converted']`), the column is removed from the dataframe. Re-running the same cell tries to delete a column that no longer exists, so pandas raises a `KeyError`.
+
+A common fix is to make the deletion idempotent by using `drop(..., errors='ignore')`, which won’t crash if the column is already gone:
+
+```python
+# Safe way to remove the target variable on multiple runs
+df_train = df_train.drop(columns=['converted'], errors='ignore')
+df_val = df_val.drop(columns=['converted'], errors='ignore')
+df_test = df_test.drop(columns=['converted'], errors='ignore')
+```
+
+If you want to avoid rerun issues altogether, you can also structure your preprocessing so the “rename/drop target” step only executes once per notebook run (or reload the original dataframe before re-running.
\ No newline at end of file