diff --git a/imap_processing/cdf/utils.py b/imap_processing/cdf/utils.py
index c854c71b6..212414db6 100644
--- a/imap_processing/cdf/utils.py
+++ b/imap_processing/cdf/utils.py
@@ -1,11 +1,15 @@
 """Various utility functions to support creation of CDF files."""
 
-import os
+import logging
+from pathlib import Path
+from typing import Optional
 
 import numpy as np
 import xarray as xr
 from cdflib.xarray import xarray_to_cdf
 
+import imap_processing
+
 
 def calc_start_time(shcoarse_time: int):
     """Calculate the datetime64 from the CCSDS secondary header information.
@@ -36,7 +40,10 @@ def calc_start_time(shcoarse_time: int):
 
 
 def write_cdf(
-    data: xr.Dataset, description: str = "", mode: str = "", directory: str = ""
+    data: xr.Dataset,
+    description: str = "",
+    mode: str = "",
+    directory: Optional[Path] = None,
 ):
     """Write the contents of "data" to a CDF file using cdflib.xarray_to_cdf.
 
@@ -57,8 +64,8 @@ def write_cdf(
 
     Returns
     -------
-        str
-            The name of the file created
+        Path
+            Path to the file created
     """
     # Determine the start date of the data in the file,
     # based on the time of the first dust impact
@@ -98,7 +105,21 @@ def write_cdf(
         + date_string
         + f"_v{data.attrs['Data_version']}.cdf"
     )
-    filename_and_path = os.path.join(directory, filename)
+
+    if directory is None:
+        # Storage directory
+        # mission/instrument/data_level/year/month/filename
+        # /<directory | DATA_DIR>/<instrument>/<data_level>/<year>/<month>
+        _, instrument, data_level = data.attrs["Logical_source"].split("_")
+        directory = imap_processing.DATA_DIR / instrument / data_level
+        directory /= date_string[:4] / date_string[4:6]
+    filename_and_path = Path(directory)
+    if not filename_and_path.exists():
+        logging.info(
+            "The directory does not exist, creating directory %s", filename_and_path
+        )
+        filename_and_path.mkdir(parents=True)
+    filename_and_path /= filename
 
     # Insert the final attribute:
     # The Logical_file_id is always the name of the file without the extension
diff --git a/imap_processing/idex/tests/test_l1_cdfs.py b/imap_processing/idex/tests/test_l1_cdfs.py
index afe9f213f..dc61620f2 100644
--- a/imap_processing/idex/tests/test_l1_cdfs.py
+++ b/imap_processing/idex/tests/test_l1_cdfs.py
@@ -1,6 +1,3 @@
-import os
-from pathlib import Path
-
 import numpy as np
 import pytest
 import xarray as xr
@@ -26,11 +23,11 @@ def test_idex_cdf_file(decom_test_data, temp_path):
     # Verify that a CDF file can be created with no errors thrown by xarray_to_cdf
     file_name = write_cdf(decom_test_data.data, description="", directory=temp_path)
     date_to_test = "20250724"
-    assert file_name == os.path.join(
-        temp_path,
-        f"{decom_test_data.data.attrs['Logical_source']}_{date_to_test}_v{idex.__version__}.cdf",
+    assert file_name.name == (
+        f"{decom_test_data.data.attrs['Logical_source']}_"
+        f"{date_to_test}_v{idex.__version__}.cdf"
     )
-    assert Path(file_name).exists()
+    assert file_name.exists()
 
 
 def test_bad_cdf_attributes(decom_test_data, temp_path):
@@ -75,11 +72,11 @@ def test_descriptor_in_file_name(decom_test_data, temp_path):
         decom_test_data.data, description="impact-lab-test001", directory=temp_path
     )
     date_to_test = "20250724"
-    assert file_name == os.path.join(
-        temp_path,
-        f"{decom_test_data.data.attrs['Logical_source']}_impact-lab-test001_{date_to_test}_v{idex.__version__}.cdf",
+    assert file_name.name == (
+        f"{decom_test_data.data.attrs['Logical_source']}_"
+        f"impact-lab-test001_{date_to_test}_v{idex.__version__}.cdf"
     )
-    assert Path(file_name).exists()
+    assert file_name.exists()
 
 
 def test_idex_tof_high_data_from_cdf(decom_test_data, temp_path):
diff --git a/imap_processing/swe/l1a/swe_l1a.py b/imap_processing/swe/l1a/swe_l1a.py
index 70c1a0cae..467f03714 100644
--- a/imap_processing/swe/l1a/swe_l1a.py
+++ b/imap_processing/swe/l1a/swe_l1a.py
@@ -28,8 +28,8 @@ def swe_l1a(packets, cdf_filepath):
 
     Returns
     -------
-    str
-        Path name of where CDF file was created.
+    pathlib.Path
+        Path to where the CDF file was created.
         This is used to upload file from local to s3.
         TODO: test this later.
     """
diff --git a/imap_processing/swe/l1b/swe_l1b.py b/imap_processing/swe/l1b/swe_l1b.py
index 118eccdb9..5a01bac6e 100644
--- a/imap_processing/swe/l1b/swe_l1b.py
+++ b/imap_processing/swe/l1b/swe_l1b.py
@@ -22,7 +22,7 @@ def swe_l1b(l1a_dataset: xr.Dataset, cdf_filepath: str):
 
     Returns
     -------
-    str
+    pathlib.Path
         Path to the L1B file.
 
     Raises
diff --git a/imap_processing/swe/tests/test_swe_l1a.py b/imap_processing/swe/tests/test_swe_l1a.py
index 85c2c7e1d..a270bdfe5 100644
--- a/imap_processing/swe/tests/test_swe_l1a.py
+++ b/imap_processing/swe/tests/test_swe_l1a.py
@@ -1,5 +1,3 @@
-import os
-
 import pytest
 
 from imap_processing import imap_module_directory
@@ -53,5 +51,5 @@ def test_cdf_creation(decom_test_data, tmp_path):
     grouped_data = group_by_apid(decom_test_data)
     sci_cdf_filepath = swe_l1a(grouped_data[SWEAPID.SWE_SCIENCE], cdf_filepath)
     hk_cdf_filepath = swe_l1a(grouped_data[SWEAPID.SWE_APP_HK], cdf_filepath)
-    assert os.path.basename(sci_cdf_filepath) == "imap_swe_l1a_sci_20230927_v01.cdf"
-    assert os.path.basename(hk_cdf_filepath) == "imap_swe_l1a_lveng_hk_20230927_v01.cdf"
+    assert sci_cdf_filepath.name == "imap_swe_l1a_sci_20230927_v01.cdf"
+    assert hk_cdf_filepath.name == "imap_swe_l1a_lveng_hk_20230927_v01.cdf"
diff --git a/imap_processing/swe/tests/test_swe_l1b.py b/imap_processing/swe/tests/test_swe_l1b.py
index fcf9855bf..99d169fa8 100644
--- a/imap_processing/swe/tests/test_swe_l1b.py
+++ b/imap_processing/swe/tests/test_swe_l1b.py
@@ -1,5 +1,3 @@
-from pathlib import Path
-
 import pandas as pd
 import pytest
 from cdflib.xarray import cdf_to_xarray
@@ -154,5 +152,5 @@ def test_cdf_creation(decom_test_data, l1a_test_data, tmp_path):
     l1a_dataset = cdf_to_xarray(hk_l1a_filepath)
     hk_l1b_filepath = swe_l1b(l1a_dataset, cdf_filepath)
 
-    assert Path(hk_l1b_filepath).name == "imap_swe_l1b_lveng_hk_20230927_v01.cdf"
-    assert Path(sci_l1b_filepath).name == "imap_swe_l1b_sci_20230927_v01.cdf"
+    assert hk_l1b_filepath.name == "imap_swe_l1b_lveng_hk_20230927_v01.cdf"
+    assert sci_l1b_filepath.name == "imap_swe_l1b_sci_20230927_v01.cdf"
diff --git a/pyproject.toml b/pyproject.toml
index 065426149..156ced512 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -72,7 +72,7 @@ select = ["B", "D", "E", "F", "I", "N", "S", "W", "PL", "PT", "UP", "RUF"]
 ignore = ["D104", "PLR2004", "S101"]
 
 [tool.ruff.per-file-ignores]
-# S603: unchecked input in subprocess call is fine in our tests
+# S603 unchecked input in subprocess call is fine in our tests
 "*/tests/*" = ["D", "S603"]
 "tools/xtce*" = ["D"]
 # TODO: Too many statements, this could be refactored and removed