diff --git a/bigframes/core/nodes.py b/bigframes/core/nodes.py
index e65040686e..d750ee63fb 100644
--- a/bigframes/core/nodes.py
+++ b/bigframes/core/nodes.py
@@ -508,6 +508,7 @@ class GbqTable:
     table_id: str = field()
     physical_schema: Tuple[bq.SchemaField, ...] = field()
     n_rows: int = field()
+    is_physical_table: bool = field()
     cluster_cols: typing.Optional[Tuple[str, ...]]
 
     @staticmethod
@@ -523,6 +524,7 @@ def from_table(table: bq.Table, columns: Sequence[str] = ()) -> GbqTable:
             table_id=table.table_id,
             physical_schema=schema,
             n_rows=table.num_rows,
+            is_physical_table=(table.table_type == "TABLE"),
             cluster_cols=None
             if table.clustering_fields is None
             else tuple(table.clustering_fields),
@@ -603,7 +605,7 @@ def variables_introduced(self) -> int:
 
     @property
     def row_count(self) -> typing.Optional[int]:
-        if self.source.sql_predicate is None:
+        if self.source.sql_predicate is None and self.source.table.is_physical_table:
             return self.source.table.n_rows
         return None
 
diff --git a/tests/system/small/test_dataframe.py b/tests/system/small/test_dataframe.py
index cfd6efe9bd..6ee9fb8247 100644
--- a/tests/system/small/test_dataframe.py
+++ b/tests/system/small/test_dataframe.py
@@ -1521,6 +1521,26 @@ def test_shape(scalars_dfs):
     assert bf_result == pd_result
 
 
+@pytest.mark.parametrize(
+    "reference_table, test_table",
+    [
+        (
+            "bigframes-dev.bigframes_tests_sys.base_table",
+            "bigframes-dev.bigframes_tests_sys.base_table_view",
+        ),
+        (
+            "bigframes-dev.bigframes_tests_sys.csv_native_table",
+            "bigframes-dev.bigframes_tests_sys.csv_external_table",
+        ),
+    ],
+)
+def test_view_and_external_table_shape(session, reference_table, test_table):
+    reference_df = session.read_gbq(reference_table)
+    test_df = session.read_gbq(test_table)
+
+    assert test_df.shape == reference_df.shape
+
+
 def test_len(scalars_dfs):
     scalars_df, scalars_pandas_df = scalars_dfs
     bf_result = len(scalars_df)