diff --git a/grand_challenge_forge/partials/example-evaluation-method/example-evaluation-method{{cookiecutter._}}/helpers.py b/grand_challenge_forge/partials/example-evaluation-method/example-evaluation-method{{cookiecutter._}}/helpers.py
index 731b47c..b69a06b 100644
--- a/grand_challenge_forge/partials/example-evaluation-method/example-evaluation-method{{cookiecutter._}}/helpers.py
+++ b/grand_challenge_forge/partials/example-evaluation-method/example-evaluation-method{{cookiecutter._}}/helpers.py
@@ -44,6 +44,8 @@ def run_prediction_processing(*, fn, predictions):
     - if any child process is terminated, all prediction processing will abort
     - after prediction processing is done, all child processes are terminated
 
+    Note that the results are returned in completing order.
+
     Parameters
     ----------
     fn : function
diff --git a/tests/test_evaluate_helpers.py b/tests/test_evaluate_helpers.py
index 20d75dd..d0856b6 100644
--- a/tests/test_evaluate_helpers.py
+++ b/tests/test_evaluate_helpers.py
@@ -31,6 +31,8 @@
 
 
 def working_process(p):
+    if p == "prediction1":
+        time.sleep(2)
     return f"{p} result"
 
 
@@ -75,7 +77,7 @@ def test_prediction_processing():
     result = run_prediction_processing(
         fn=working_process, predictions=predictions
     )
-    assert ["prediction1 result", "prediction2 result"] == result
+    assert {"prediction1 result", "prediction2 result"} == set(result)
 
 
 def test_prediction_processing_error():