diff --git a/data_analysis/02-3_oracle_workflow_test.py b/data_analysis/02-3_oracle_workflow_test.py
index 6c783e8..82eec9c 100644
--- a/data_analysis/02-3_oracle_workflow_test.py
+++ b/data_analysis/02-3_oracle_workflow_test.py
@@ -1,4 +1,5 @@
 # %%
+import json
 import time
 from collections.abc import Sequence
 from pathlib import Path
@@ -165,8 +166,7 @@ print(stmt.compile(engine))
 
 # %%
 # raw data query
-# TODO change to left join, otherwise possible that requests are missed
-# TODO after that: look for entries which do not have an associated title number
+# TODO look for entries which do not have an associated title number
 
 print("--------------- ext_bedpbed --------------")
 t1 = time.perf_counter()
@@ -199,7 +199,50 @@ elapsed = t2 - t1
 print(f"Query duration: {elapsed:.4f} sec")
 print("Number of entries: ", len(df))
 print(f"Estimated size in memory: {df.estimated_size(unit='mb')} MB")
+# %%
+# SAVING/LOADING
+p_save = Path.cwd() / "raw_data_from_sql_query_20251202-2.arrow"
+# df.write_ipc(p_save)
+df = pl.read_ipc(p_save)
+# %%
+len(df)
+df.head()
+# 4591588: in title database with different MANDANT (are MANDANTFUEHR and BEDP_MAN feasible for matching?)
+# %%
+df.filter(pl.col("BEDP_MAN").is_in((1, 90))).filter(pl.col("MELDENUMMER"))
+# %%
+# !! CHECK: null values set in the query with CASE statement
+print(len(df.filter(pl.col("MELDENUMMER") == 18)))
+# df.filter(pl.col("MELDENUMMER") == 18).filter((pl.col("BEDP_MENGE_BEDARF_VM").is_not_null()) & (pl.col("BEDP_MENGE_BEDARF_VM") > 0))
+df.filter(pl.col("BEDP_MENGE_BEDARF_VM") > pl.col("MENGE_VORMERKER"))
+# %%
+# !! CHECK: titles with request where no title information is found
+# not_in_title_table = df.filter(pl.col("BEDP_MAN").is_in((1, 90))).filter(
+#     pl.col("MELDENUMMER").is_null()
+# )
+# EXPORT_FEAT = "BEDP_TITELNR"
+# to_save = {EXPORT_FEAT: not_in_title_table.select(EXPORT_FEAT).to_series().to_list()}
+# p_save_not_in_title_table = Path.cwd() / "not_in_title_table.json"
 
+# with open(p_save_not_in_title_table, "w") as file:
+#     json.dump(to_save, file, indent=4)
+# %%
+# !! CHECK: different MANDANTEN
+# check for valid entries for unknown MANDANTEN
+# MANDANT = 80
+
+# print(f"Mandant: {MANDANT}")
+# print(
+#     df.filter(pl.col("BEDP_MAN") == MANDANT).select(
+#         ["BEDP_MENGE_BEDARF_VM", "MELDENUMMER", "MENGE_VORMERKER"]
+#     )
+# )
+# print(
+#     df.filter(pl.col("BEDP_MAN") == MANDANT).select(
+#         ["BEDP_MENGE_BEDARF_VM", "MELDENUMMER", "MENGE_VORMERKER"]
+#     ).null_count()
+# )
+# print("Unique value counts: ", df.select(pl.col("BEDP_MAN").value_counts()))
 # %%
 # VM_CRITERION = "MENGE_VORMERKER"
 VM_CRITERION = "BEDP_MENGE_BEDARF_VM"
@@ -221,7 +264,7 @@ def get_raw_data() -> pl.DataFrame:
         ).label("BEDP_MENGE_BEDARF_VM"),
         db.ext_titel_info.c.MELDENUMMER,
         db.ext_titel_info.c.MENGE_VORMERKER,
-    ).select_from(db.ext_bedpbed.join(db.ext_titel_info, join_condition))
+    ).select_from(db.ext_bedpbed.join(db.ext_titel_info, join_condition, isouter=True))
 
     return pl.read_database(
         stmt,
diff --git a/data_analysis/not_in_title_table.json b/data_analysis/not_in_title_table.json
new file mode 100644
index 0000000..52a7c12
--- /dev/null
+++ b/data_analysis/not_in_title_table.json
@@ -0,0 +1,49 @@
+{
+    "BEDP_TITELNR": [
+        6132326,
+        4836777,
+        4836779,
+        3370676,
+        3370678,
+        6261428,
+        6261430,
+        8254295,
+        8139588,
+        6178366,
+        6178367,
+        8139587,
+        6178370,
+        6178371,
+        8139586,
+        8139585,
+        4837536,
+        3369003,
+        6132318,
+        6132319,
+        8254301,
+        6132322,
+        6132323,
+        4838000,
+        4838001,
+        4836769,
+        4836770,
+        8139590,
+        8139591,
+        3369002,
+        4837537,
+        3408130,
+        3408132,
+        5227666,
+        5227665,
+        5227663,
+        5227661,
+        139058,
+        9126790,
+        5917263,
+        7112355,
+        1462793,
+        1216207,
+        507075,
+        8254294
+    ]
+}
\ No newline at end of file
diff --git a/data_analysis/queries.sql b/data_analysis/queries.sql
new file mode 100644
index 0000000..19e1ca3
--- /dev/null
+++ b/data_analysis/queries.sql
@@ -0,0 +1,39 @@
+-- SELECT count(*) FROM EXT_TITEL_INFO
+-- SELECT * FROM EXT_TITEL_INFO
+set timing on
+
+SELECT count(*) FROM EXT_BEDPBED;
+
+-- PROMPT No Hashing allowed...
+-- SELECT /*+ NO_USE_HASH(bedp t_info) */
+-- bedp.BEDARFNR,
+-- bedp.BEDP_SEQUENZ,
+-- bedp.BEDP_TITELNR,
+-- bedp.BEDP_MAN,
+-- bedp.BEDP_MENGE_BEDARF_VM,
+-- t_info.MELDENUMMER,
+-- t_info.MENGE_VORMERKER
+-- FROM EXT_BEDPBED bedp
+-- LEFT JOIN EXT_TITEL_INFO t_info
+--     ON bedp.BEDP_TITELNR = t_info.TI_NUMMER
+--     AND bedp.BEDP_MAN = t_info.MANDFUEHR;
+-- PROMPT ####################################
+
+-- PROMPT All allowed
+-- SELECT
+-- bedp.BEDARFNR,
+-- bedp.BEDP_SEQUENZ,
+-- bedp.BEDP_TITELNR,
+-- bedp.BEDP_MAN,
+-- bedp.BEDP_MENGE_BEDARF_VM,
+-- t_info.MELDENUMMER,
+-- t_info.MENGE_VORMERKER
+-- FROM EXT_BEDPBED bedp
+-- LEFT JOIN EXT_TITEL_INFO t_info
+--     ON bedp.BEDP_TITELNR = t_info.TI_NUMMER
+--     AND bedp.BEDP_MAN = t_info.MANDFUEHR;
+-- -- WHERE bedp.BEDP_MAN IN (1, 90) AND t_info.MELDENUMMER != 26;
+-- PROMPT ######################################
+
+-- SELECT * FROM EXT_TITEL_INFO t_info WHERE t_info.TI_NUMMER = 6132326;
+SELECT * FROM EXT_TITEL_INFO t_info WHERE t_info.TI_NUMMER = 4591588;
diff --git a/data_analysis/raw_data_from_sql_query_20251202-1.arrow b/data_analysis/raw_data_from_sql_query_20251202-1.arrow
new file mode 100644
index 0000000..c77c22e
Binary files /dev/null and b/data_analysis/raw_data_from_sql_query_20251202-1.arrow differ
diff --git a/data_analysis/raw_data_from_sql_query_20251202-2.arrow b/data_analysis/raw_data_from_sql_query_20251202-2.arrow
new file mode 100644
index 0000000..f1e04ae
Binary files /dev/null and b/data_analysis/raw_data_from_sql_query_20251202-2.arrow differ