diff --git a/app/evaluation_tests.py b/app/evaluation_tests.py
index 21b8609..a9c3d45 100755
--- a/app/evaluation_tests.py
+++ b/app/evaluation_tests.py
@@ -94,6 +94,20 @@ def test_physical_qualities_no_tolerance(self):
         result = evaluation_function(response, answer, params)
         assert result["is_correct"] is False
 
+    def test_physical_qualities_correct_feedback(self):
+        params = {
+            "atol": 0.0,
+            "rtol": 0.0,
+            "strict_syntax": False,
+            "physical_quantity": True,
+            "elementary_functions": True,
+        }
+        response = "0.6 Nm"
+        answer = "0.6 Nm"
+        result = evaluation_function(response, answer, params)
+        assert result["is_correct"] is True
+        assert result["feedback"] == "Response matches answer."
+
     def test_euler_preview_evaluate(self):
         response = "ER_2"
         params = Params(is_latex=True, elementary_functions=False, strict_syntax=False)
diff --git a/app/tests/example_tests.py b/app/tests/example_tests.py
index 4e99325..1b2301d 100644
--- a/app/tests/example_tests.py
+++ b/app/tests/example_tests.py
@@ -651,8 +651,8 @@ def test_custom_comparison_with_criteria_contains(self, response, value, tags):
                     "2+answer > response_UNKNOWN",
                 ],
                 {
-                    "answer <= response_TRUE": "AAA",
-                    "2+answer > response_UNKNOWN": "BBB",
+                    "answer <= response_TRUE": "Custom response true",
+                    "2+answer > response_UNKNOWN": "μ Custom response unknown",
                 },
                 {
                     "symbol_assumptions": "('x', 'real')",
diff --git a/app/utility/evaluation_result_utilities.py b/app/utility/evaluation_result_utilities.py
index f1c0eb5..c7c222f 100644
--- a/app/utility/evaluation_result_utilities.py
+++ b/app/utility/evaluation_result_utilities.py
@@ -38,6 +38,10 @@ def add_feedback_from_tags(self, tags, graph, custom_feedback=None):
                         feedback_string = graph.criteria[tag].feedback_string_generator(dict())
                     else:
                         feedback_string = graph.criteria[tag].feedback_string_generator(inputs)
+
+                if feedback_string is not None and feedback_string[0].isalpha() and feedback_string[0].isascii():
+                    feedback_string = feedback_string.capitalize()
+
                 self.add_feedback((tag, feedback_string))
 
     def add_criteria_graph(self, name, graph):