diff --git a/sync_manager.py b/sync_manager.py
index 5dc97f9a..f3b2c9d5 100644
--- a/sync_manager.py
+++ b/sync_manager.py
@@ -83,9 +83,10 @@ class SyncManager:
         self.smart_merge_cols = ["CRM Website"]
 
     def _load_data(self):
-        # (Diese Funktion bleibt exakt wie in der letzten Version)
+        """Lädt und bereitet die Daten aus D365 und Google Sheets vor."""
         self.logger.info(f"Lade Daten aus D365-Export: '{self.d365_export_path}'...")
         try:
+            # D365 wird bereits typsicher als String geladen, das ist korrekt.
             temp_d365_df = pd.read_excel(self.d365_export_path, dtype=str).fillna('')
             for d365_col in self.d365_to_gsheet_map.keys():
                 if d365_col not in temp_d365_df.columns:
@@ -100,19 +101,29 @@ class SyncManager:
 
         self.logger.info("Lade bestehende Daten aus dem Google Sheet...")
         try:
-            all_data_with_headers = self.sheet_handler.get_all_data_with_headers()
+            # --- TYPSICHeres LADEN AUS GSHEET ---
+            # 1. Lade alle Werte mit der Option, sie als rohen TEXT zu formatieren.
+            all_data_with_headers = self.sheet_handler.sheet.get_all_values(value_render_option='FORMATTED_STRING')
+            
             if not all_data_with_headers or len(all_data_with_headers) < self.sheet_handler._header_rows:
                 self.gsheet_df = pd.DataFrame(columns=COLUMN_ORDER).fillna('')
             else:
                 actual_header = all_data_with_headers[self.sheet_handler._header_rows - 1]
                 data_rows = all_data_with_headers[self.sheet_handler._header_rows:]
+                
                 temp_df = pd.DataFrame(data_rows)
                 if not temp_df.empty:
-                    if temp_df.shape[1] > len(actual_header):
-                        temp_df = temp_df.iloc[:, :len(actual_header)]
+                    # Spaltenanzahl an Header anpassen
+                    num_cols_to_match = len(actual_header)
+                    if temp_df.shape[1] > num_cols_to_match:
+                        temp_df = temp_df.iloc[:, :num_cols_to_match]
+                    elif temp_df.shape[1] < num_cols_to_match:
+                        for i in range(num_cols_to_match - temp_df.shape[1]):
+                            temp_df[f'temp_{i}'] = ''
                     temp_df.columns = actual_header
                 else:
                     temp_df = pd.DataFrame(columns=actual_header)
+                
                 temp_df = temp_df.fillna('')
                 for col_name in COLUMN_ORDER:
                     if col_name not in temp_df.columns: