diff --git a/tests/unit/aiplatform/test_endpoints.py b/tests/unit/aiplatform/test_endpoints.py
index 21a55b5581..09524fd23a 100644
--- a/tests/unit/aiplatform/test_endpoints.py
+++ b/tests/unit/aiplatform/test_endpoints.py
@@ -107,6 +107,9 @@
 _TEST_ACCELERATOR_TYPE = "NVIDIA_TESLA_P100"
 _TEST_ACCELERATOR_COUNT = 2
 
+_TEST_METRIC_NAME_CPU_UTILIZATION = "aiplatform.googleapis.com/prediction/online/cpu/utilization"
+_TEST_METRIC_NAME_GPU_UTILIZATION = "aiplatform.googleapis.com/prediction/online/accelerator/duty_cycle"
+
 _TEST_EXPLANATIONS = [gca_prediction_service.explanation.Explanation(attributions=[])]
 
 _TEST_ATTRIBUTIONS = [
@@ -1082,8 +1085,8 @@ def test_deploy_with_autoscaling_target_cpu_utilization(self, deploy_model_mock,
         )
 
         expected_autoscaling_metric_spec = gca_machine_resources.AutoscalingMetricSpec(
-            metric_name="aiplatform.googleapis.com/prediction/online/cpu/utilization",
-            target=70
+            metric_name=_TEST_METRIC_NAME_CPU_UTILIZATION,
+            target=70,
         )
 
         expected_dedicated_resources = gca_machine_resources.DedicatedResources(
@@ -1136,8 +1139,8 @@ def test_deploy_with_autoscaling_target_accelerator_duty_cycle(self, deploy_mode
         )
 
         expected_autoscaling_metric_spec = gca_machine_resources.AutoscalingMetricSpec(
-            metric_name="aiplatform.googleapis.com/prediction/online/accelerator/duty_cycle",
-            target=70
+            metric_name=_TEST_METRIC_NAME_GPU_UTILIZATION,
+            target=70,
         )
 
         expected_dedicated_resources = gca_machine_resources.DedicatedResources(