diff --git a/.github/workflows/generate_reference_results_workflow.yml b/.github/workflows/generate_reference_results_workflow.yml
index f3677612d..4b647ba97 100644
--- a/.github/workflows/generate_reference_results_workflow.yml
+++ b/.github/workflows/generate_reference_results_workflow.yml
@@ -64,7 +64,7 @@ jobs:
         git checkout ${{ inputs.from_ref }}
         git add ./*/*/*.tar.gz
         git add ./*/*.tar.gz
-        git add ./*/*.metadata
+        git add ./*/reference-results/*.metadata
         git commit -m "${{inputs.commit_msg}}"
         git push
     - name: Upload artifacts for debugging
diff --git a/changelog-entries/537.md b/changelog-entries/537.md
new file mode 100644
index 000000000..9adfb27db
--- /dev/null
+++ b/changelog-entries/537.md
@@ -0,0 +1 @@
+- Moved `reference_results.metadata` into the `reference-results/` directory for each tutorial, colocating it with the reference result archives it describes. #537
diff --git a/elastic-tube-1d/reference_results.metadata b/elastic-tube-1d/reference-results/reference_results.metadata
similarity index 100%
rename from elastic-tube-1d/reference_results.metadata
rename to elastic-tube-1d/reference-results/reference_results.metadata
diff --git a/flow-over-heated-plate/reference_results.metadata b/flow-over-heated-plate/reference-results/reference_results.metadata
similarity index 100%
rename from flow-over-heated-plate/reference_results.metadata
rename to flow-over-heated-plate/reference-results/reference_results.metadata
diff --git a/multiple-perpendicular-flaps/reference_results.metadata b/multiple-perpendicular-flaps/reference-results/reference_results.metadata
similarity index 100%
rename from multiple-perpendicular-flaps/reference_results.metadata
rename to multiple-perpendicular-flaps/reference-results/reference_results.metadata
diff --git a/perpendicular-flap/reference_results.metadata b/perpendicular-flap/reference-results/reference_results.metadata
similarity index 100%
rename from perpendicular-flap/reference_results.metadata
rename to perpendicular-flap/reference-results/reference_results.metadata
diff --git a/tools/tests/generate_reference_results.py b/tools/tests/generate_reference_results.py
index f43866745..055e7b31c 100644
--- a/tools/tests/generate_reference_results.py
+++ b/tools/tests/generate_reference_results.py
@@ -1,4 +1,3 @@
-
 import argparse
 from metadata_parser.metdata import Tutorials, ReferenceResult
 from systemtests.TestSuite import TestSuites
@@ -142,7 +141,9 @@ def main():
 
     # write readme
     for tutorial in reference_result_per_tutorial.keys():
-        with open(tutorial.path / "reference_results.metadata", 'w') as file:
+        reference_results_dir = tutorial.path / "reference-results"
+        reference_results_dir.mkdir(parents=True, exist_ok=True)
+        with open(reference_results_dir / "reference_results.metadata", 'w') as file:
             ref_results_info = render_reference_results_info(
                 reference_result_per_tutorial[tutorial], build_args, current_time_string)
             logging.info(f"Writing results for {tutorial.name}")