diff --git a/docs/changelog.rst b/docs/changelog.rst
index cb31da21..d8765d17 100644
--- a/docs/changelog.rst
+++ b/docs/changelog.rst
@@ -10,7 +10,9 @@ dev
 -  Fix validation error instance in GenericEmbeddedDocumentField #1067
 -  Update cached fields when fields argument is given #1712
 -  Add a db parameter to register_connection for compatibility with connect
+-  Use insert_one, insert_many in Document.insert #1491
 -  Use new update_one, update_many on document/queryset update #1491
+-  Use insert_one, insert_many in Document.insert #1491
 
 Changes in 0.15.0
 =================
diff --git a/mongoengine/queryset/base.py b/mongoengine/queryset/base.py
index f78ee882..647eafc6 100644
--- a/mongoengine/queryset/base.py
+++ b/mongoengine/queryset/base.py
@@ -350,11 +350,24 @@ class BaseQuerySet(object):
                                      documents=docs, **signal_kwargs)
 
         raw = [doc.to_mongo() for doc in docs]
+
+        with set_write_concern(self._collection, write_concern) as collection:
+            insert_func = collection.insert_many
+            if return_one:
+                raw = raw[0]
+                insert_func = collection.insert_one
+
         try:
-            ids = self._collection.insert(raw, **write_concern)
+            inserted_result = insert_func(raw)
+            ids = return_one and [inserted_result.inserted_id] or inserted_result.inserted_ids
         except pymongo.errors.DuplicateKeyError as err:
             message = 'Could not save document (%s)'
             raise NotUniqueError(message % six.text_type(err))
+        except pymongo.errors.BulkWriteError as err:
+            # inserting documents that already have an _id field will
+            # give huge performance debt or raise
+            message = u'Document must not have _id value before bulk write (%s)'
+            raise NotUniqueError(message % six.text_type(err))
         except pymongo.errors.OperationFailure as err:
             message = 'Could not save document (%s)'
             if re.match('^E1100[01] duplicate key', six.text_type(err)):
@@ -368,7 +381,6 @@ class BaseQuerySet(object):
             signals.post_bulk_insert.send(
                 self._document, documents=docs, loaded=False, **signal_kwargs)
             return return_one and ids[0] or ids
-
         documents = self.in_bulk(ids)
         results = []
         for obj_id in ids:
diff --git a/tests/queryset/queryset.py b/tests/queryset/queryset.py
index 5c3f179f..3fa86a58 100644
--- a/tests/queryset/queryset.py
+++ b/tests/queryset/queryset.py
@@ -927,10 +927,6 @@ class QuerySetTest(unittest.TestCase):
 
         self.assertEqual(Blog.objects.count(), 2)
 
-        Blog.objects.insert([blog2, blog3],
-                            write_concern={"w": 0, 'continue_on_error': True})
-        self.assertEqual(Blog.objects.count(), 3)
-
     def test_get_changed_fields_query_count(self):
         """Make sure we don't perform unnecessary db operations when
         none of document's fields were updated.