diff --git a/it/list_test.py b/it/list_test.py
index 24339ac7..03ad6569 100644
--- a/it/list_test.py
+++ b/it/list_test.py
@@ -27,7 +27,7 @@
 
 @it.all_benchmark_configs
 def test_list_test_runs(cfg):
-    assert it.osbenchmark(cfg, "list test_runs") == 0
+    assert it.osbenchmark(cfg, "list test-runs") == 0
 
 
 @it.benchmark_in_mem
diff --git a/osbenchmark/benchmark.py b/osbenchmark/benchmark.py
index cf0cc016..d31a755b 100644
--- a/osbenchmark/benchmark.py
+++ b/osbenchmark/benchmark.py
@@ -201,12 +201,12 @@ def add_workload_source(subparser):
         "--baseline",
         "-b",
         required=True,
-        help=f"TestRun ID of the baseline (see {PROGRAM_NAME} list test_runs).")
+        help=f"TestRun ID of the baseline (see {PROGRAM_NAME} list test-runs).")
     compare_parser.add_argument(
         "--contender",
         "-c",
         required=True,
-        help=f"TestRun ID of the contender (see {PROGRAM_NAME} list test_runs).")
+        help=f"TestRun ID of the contender (see {PROGRAM_NAME} list test-runs).")
     compare_parser.add_argument(
         "--percentiles",
         help=f"A comma-separated list of percentiles to report latency and service time."