diff --git a/parsers/common/cellparser.py b/parsers/common/cellparser.py
index 0d01bb7..3c9a776 100644
--- a/parsers/common/cellparser.py
+++ b/parsers/common/cellparser.py
@@ -68,16 +68,18 @@ def parse(self, value, context={}):
             return self.split_into_lists(value)
 
     def parse_as_string(self, value, context={}, is_object=None):
+        # If context is None, template parsing is omitted entirely.
         # is_object is a pass-by-reference boolean, realised via
         # the class BooleanWrapper, to indicate to the caller
         # whether the parsing result represents an object that
         # is not to be processed any further.
         if value is None:
             return ''
-        if not context and '{' not in value:
+        if context is None or (not context and '{' not in value):
             # This is a hacky optimization.
             return value
         stripped_value = value.strip()
+        env = self.env
         if stripped_value.startswith('{@') and stripped_value.endswith('@}'):
             # Special case: Return a python object rather than a string,
             # if possible.
@@ -85,8 +87,10 @@ def parse_as_string(self, value, context={}, is_object=None):
             assert stripped_value[2:].find('{@') == -1
             if is_object is not None:
                 is_object.boolean = True
-            template = self.native_env.from_string(stripped_value)
-            return template.render(context)
-        else:
-            template = self.env.from_string(stripped_value)
+            env = self.native_env
+
+        try:
+            template = env.from_string(stripped_value)
             return template.render(context)
+        except Exception as e:
+            raise ValueError(str(e), stripped_value, context)
diff --git a/parsers/common/sheetparser.py b/parsers/common/sheetparser.py
index 0e20b7e..9317146 100644
--- a/parsers/common/sheetparser.py
+++ b/parsers/common/sheetparser.py
@@ -35,12 +35,13 @@ def go_to_bookmark(self, name):
     def remove_bookmark(self, name):
         self.bookmarks.pop(name)
 
-    def parse_next_row(self):
+    def parse_next_row(self, omit_templating=False):
         try:
             input_row = next(self.iterator)
         except StopIteration:
             return None
-        row = self.row_parser.parse_row(input_row, self.context)
+        context = self.context if not omit_templating else None
+        row = self.row_parser.parse_row(input_row, context)
         return row
 
     def parse_all(self):
diff --git a/parsers/common/tests/mock_sheetparser.py b/parsers/common/tests/mock_sheetparser.py
index b35d979..af7686e 100644
--- a/parsers/common/tests/mock_sheetparser.py
+++ b/parsers/common/tests/mock_sheetparser.py
@@ -17,7 +17,7 @@ def __init__(self, row_parser, rows, context={}):
         self.iterator = iter(self.input_rows)
         self.context = copy.deepcopy(context)
 
-    def parse_next_row(self):
+    def parse_next_row(self, omit_templating=False):
         # Simply return the input.
         try:
             input_row = next(self.iterator)
diff --git a/parsers/creation/contentindexparser.py b/parsers/creation/contentindexparser.py
index 11f2cc6..6a0d7e9 100644
--- a/parsers/creation/contentindexparser.py
+++ b/parsers/creation/contentindexparser.py
@@ -142,8 +142,8 @@ def map_template_arguments_to_context(self, arg_defs, args, context):
 		for arg_def, arg in zip(arg_defs, args + args_padding):
 			if arg_def.name in context:
 				raise ValueError(f'Template argument "{arg_def.name}" doubly defined in context')
-			arg_value = arg or arg_def.default_value
-			if not arg_value:
+			arg_value = arg if arg != '' else arg_def.default_value
+			if arg_value == '':
 				raise ValueError(f'Required template argument "{arg_def.name}" not provided')
 			if arg_def.type == 'sheet':
 				context[arg_def.name] = self.data_sheets[arg_value]
diff --git a/parsers/creation/flowparser.py b/parsers/creation/flowparser.py
index 3e5a41e..588154d 100644
--- a/parsers/creation/flowparser.py
+++ b/parsers/creation/flowparser.py
@@ -304,7 +304,7 @@ def parse(self):
         return flow_container
 
     def _parse_block(self, depth=0, block_type='root_block', omit_content=False):
-        row = self.sheet_parser.parse_next_row()
+        row = self.sheet_parser.parse_next_row(omit_templating=omit_content)
         while not self._is_end_of_block(block_type, row):
             if omit_content or not row.include_if:
                 if row.type == 'begin_for':
@@ -349,7 +349,7 @@ def _parse_block(self, depth=0, block_type='root_block', omit_content=False):
                     self.append_node_group(new_node_group, row.row_id)
                 else:
                     self._parse_row(row)
-            row = self.sheet_parser.parse_next_row()
+            row = self.sheet_parser.parse_next_row(omit_templating=omit_content)
 
     def _is_end_of_block(self, block_type, row):
         block_end_map = {