diff --git a/heracles/io.py b/heracles/io.py
index 2f88a40..d5a6e3d 100644
--- a/heracles/io.py
+++ b/heracles/io.py
@@ -61,6 +61,37 @@ def _read_metadata(hdu):
     return md
 
 
+def _as_twopoint(arr, name):
+    """convert two-point data (i.e. one L column) to structured array"""
+
+    arr = np.asanyarray(arr)
+
+    # get the data into structured array if not already
+    if arr.dtype.names is None:
+        n, *dims = arr.shape
+        data = arr
+
+        dt = np.dtype(
+            [
+                ("L", float),
+                (name, arr.dtype.str, dims) if dims else (name, arr.dtype.str),
+                ("LMIN", float),
+                ("LMAX", float),
+                ("W", float),
+            ],
+            metadata=dict(arr.dtype.metadata or {}),
+        )
+
+        arr = np.empty(n, dt)
+        arr["L"] = np.arange(n)
+        arr[name] = data
+        arr["LMIN"] = arr["L"]
+        arr["LMAX"] = arr["L"] + 1
+        arr["W"] = 1
+
+    return arr
+
+
 def read_mask(mask_name, nside=None, field=0, extra_mask_name=None):
     """read visibility map from a HEALPix map file"""
     mask = hp.read_map(mask_name, field=field)
@@ -397,27 +428,10 @@ def write_cls(filename, cls, *, clobber=False, workdir=".", include=None, exclud
             ext = f"CL{cln}"
             cln += 1
 
-            # get the data into the binned format if not already
-            if cl.dtype.names is None:
-                dt = np.dtype(
-                    [
-                        ("L", float),
-                        ("CL", float),
-                        ("LMIN", float),
-                        ("LMAX", float),
-                        ("W", float),
-                    ],
-                    metadata=dict(cl.dtype.metadata or {}),
-                )
-                cl_ = cl
-                cl = np.empty(len(cl_), dt)
-                cl["L"] = np.arange(len(cl_))
-                cl["CL"] = cl_
-                cl["LMIN"] = cl["L"]
-                cl["LMAX"] = cl["L"] + 1
-                cl["W"] = 1
-
-            # write the data column
+            # get the data into structured format if not already
+            cl = _as_twopoint(cl, "CL")
+
+            # write the data columns
             fits.write_table(cl, extname=ext)
 
             # write the metadata
@@ -519,17 +533,11 @@ def write_mms(filename, mms, *, clobber=False, workdir=".", include=None, exclud
             ext = f"MM{mmn}"
             mmn += 1
 
-            # write the mixing matrix as an image
-            fits.write_image(mm, extname=ext)
+            # get the data into structured format if not already
+            mm = _as_twopoint(mm, "MM")
 
-            # write the WCS
-            fits[ext].write_key("WCSAXES", 2)
-            fits[ext].write_key("CNAME1", "L_1")
-            fits[ext].write_key("CNAME2", "L_2")
-            fits[ext].write_key("CTYPE1", " ")
-            fits[ext].write_key("CTYPE2", " ")
-            fits[ext].write_key("CUNIT1", " ")
-            fits[ext].write_key("CUNIT2", " ")
+            # write the data columns
+            fits.write_table(mm, extname=ext)
 
             # write the metadata
             _write_metadata(fits[ext], mm.dtype.metadata)
diff --git a/tests/test_io.py b/tests/test_io.py
index d98d1c4..7939455 100644
--- a/tests/test_io.py
+++ b/tests/test_io.py
@@ -276,7 +276,11 @@ def test_write_read_mms(tmp_path):
 
     assert mms_.keys() == mms.keys()
     for key in mms:
-        np.testing.assert_array_equal(mms_[key], mms[key])
+        assert key in mms_
+        mm = mms_[key]
+        assert mm.dtype.names == ("L", "MM", "LMIN", "LMAX", "W")
+        np.testing.assert_array_equal(mm["L"], np.arange(len(mms[key])))
+        np.testing.assert_array_equal(mm["MM"], mms[key])
 
 
 def test_write_read_cov(mock_cls, tmp_path):