diff --git a/awswrangler/redshift/_utils.py b/awswrangler/redshift/_utils.py
index b3d05aebe..b5409cd02 100644
--- a/awswrangler/redshift/_utils.py
+++ b/awswrangler/redshift/_utils.py
@@ -379,31 +379,6 @@ def _create_table(  # noqa: PLR0913
     add_new_columns: bool = False,
 ) -> tuple[str, str | None]:
     _logger.debug("Creating table %s with mode %s, and overwrite method %s", table, mode, overwrite_method)
-    redshift_types = _get_rsh_types(
-        df=df,
-        path=path,
-        index=index,
-        dtype=dtype,
-        varchar_lengths_default=varchar_lengths_default,
-        varchar_lengths=varchar_lengths,
-        parquet_infer_sampling=parquet_infer_sampling,
-        path_suffix=path_suffix,
-        path_ignore_suffix=path_ignore_suffix,
-        use_threads=use_threads,
-        boto3_session=boto3_session,
-        s3_additional_kwargs=s3_additional_kwargs,
-        data_format=data_format,
-        redshift_column_types=redshift_column_types,
-        manifest=manifest,
-    )
-    if add_new_columns is True:
-        if _does_table_exist(cursor=cursor, schema=schema, table=table) is True:
-            actual_table_columns = set(_get_table_columns(cursor=cursor, schema=schema, table=table))
-            new_df_columns = {
-                key: value for key, value in redshift_types.items() if key.lower() not in actual_table_columns
-            }
-            _add_table_columns(cursor=cursor, schema=schema, table=table, new_columns=new_df_columns)
-
     if mode == "overwrite":
         if overwrite_method == "truncate":
             try:
@@ -433,6 +408,29 @@ def _create_table(  # noqa: PLR0913
         _logger.debug("Table %s exists", table)
         if lock:
             _lock(cursor, [table], schema=schema)
+        if add_new_columns is True:
+            redshift_types = _get_rsh_types(
+                df=df,
+                path=path,
+                index=index,
+                dtype=dtype,
+                varchar_lengths_default=varchar_lengths_default,
+                varchar_lengths=varchar_lengths,
+                parquet_infer_sampling=parquet_infer_sampling,
+                path_suffix=path_suffix,
+                path_ignore_suffix=path_ignore_suffix,
+                use_threads=use_threads,
+                boto3_session=boto3_session,
+                s3_additional_kwargs=s3_additional_kwargs,
+                data_format=data_format,
+                redshift_column_types=redshift_column_types,
+                manifest=manifest,
+            )
+            actual_table_columns = set(_get_table_columns(cursor=cursor, schema=schema, table=table))
+            new_df_columns = {
+                key: value for key, value in redshift_types.items() if key.lower() not in actual_table_columns
+            }
+            _add_table_columns(cursor=cursor, schema=schema, table=table, new_columns=new_df_columns)
         if mode == "upsert":
             guid: str = uuid.uuid4().hex
             temp_table: str = f"temp_redshift_{guid}"
@@ -444,6 +442,23 @@ def _create_table(  # noqa: PLR0913
     diststyle = diststyle.upper() if diststyle else "AUTO"
     sortstyle = sortstyle.upper() if sortstyle else "COMPOUND"
 
+    redshift_types = _get_rsh_types(
+        df=df,
+        path=path,
+        index=index,
+        dtype=dtype,
+        varchar_lengths_default=varchar_lengths_default,
+        varchar_lengths=varchar_lengths,
+        parquet_infer_sampling=parquet_infer_sampling,
+        path_suffix=path_suffix,
+        path_ignore_suffix=path_ignore_suffix,
+        use_threads=use_threads,
+        boto3_session=boto3_session,
+        s3_additional_kwargs=s3_additional_kwargs,
+        data_format=data_format,
+        redshift_column_types=redshift_column_types,
+        manifest=manifest,
+    )
     _validate_parameters(
         redshift_types=redshift_types,
         diststyle=diststyle,