diff --git a/ibis/formats/pandas.py b/ibis/formats/pandas.py
index b2efdd21a7349..ea67c9ad75378 100644
--- a/ibis/formats/pandas.py
+++ b/ibis/formats/pandas.py
@@ -109,14 +109,11 @@ def infer_table(cls, df):
 
     @classmethod
     def convert_table(cls, df, schema):
-        if len(schema) != len(df.columns):
-            raise ValueError(
-                "schema column count does not match input data column count"
-            )
+        if schema.names != tuple(df.columns):
+            raise ValueError("schema names don't match input data columns")
 
         columns = {
-            name: cls.convert_column(series, dtype)
-            for (name, dtype), (_, series) in zip(schema.items(), df.items())
+            name: cls.convert_column(df[name], dtype) for name, dtype in schema.items()
         }
         df = pd.DataFrame(columns)
 
diff --git a/ibis/formats/tests/test_pandas.py b/ibis/formats/tests/test_pandas.py
index 671d3be1b99c5..13c33da6bcfea 100644
--- a/ibis/formats/tests/test_pandas.py
+++ b/ibis/formats/tests/test_pandas.py
@@ -433,3 +433,10 @@ def test_convert_dataframe_with_timezone():
     desired_schema = ibis.schema(dict(time='timestamp("EST")'))
     result = PandasData.convert_table(df.copy(), desired_schema)
     tm.assert_frame_equal(expected, result)
+
+
+def test_schema_doesnt_match_input_columns():
+    df = pd.DataFrame({"x": [1], "y": [2]})
+    schema = sch.Schema({"a": "int64", "b": "int64"})
+    with pytest.raises(ValueError, match="schema names don't match"):
+        PandasData.convert_table(df, schema)