diff --git a/minari/data_collector/data_collector.py b/minari/data_collector/data_collector.py
index f380b915..a3f20135 100644
--- a/minari/data_collector/data_collector.py
+++ b/minari/data_collector/data_collector.py
@@ -289,10 +289,7 @@ def create_dataset(
         )
 
         self._save_to_disk(dataset_path, metadata)
-
-        dataset = MinariDataset(dataset_path)
-        dataset.storage.update_metadata({"dataset_size": dataset.storage.get_size()})
-        return dataset
+        return MinariDataset(dataset_path)
 
     def _flush_to_storage(self):
         if self._buffer is not None and len(self._buffer) > 0:
diff --git a/minari/dataset/_storages/arrow_storage.py b/minari/dataset/_storages/arrow_storage.py
index 0cc0530c..b3582c11 100644
--- a/minari/dataset/_storages/arrow_storage.py
+++ b/minari/dataset/_storages/arrow_storage.py
@@ -65,6 +65,8 @@ def update_episode_metadata(
             with open(metadata_path, "w") as file:
                 json.dump(metadata, file, cls=NumpyEncoder)
 
+        self.update_metadata({"dataset_size": self.get_size()})
+
     def get_episode_metadata(self, episode_indices: Iterable[int]) -> Iterable[Dict]:
         for episode_id in episode_indices:
             metadata_path = self.data_path.joinpath(str(episode_id), "metadata.json")
@@ -148,7 +150,11 @@ def update_episodes(self, episodes: Iterable[EpisodeBuffer]):
             self.update_episode_metadata([episode_metadata], [episode_id])
 
         self.update_metadata(
-            {"total_steps": total_steps, "total_episodes": total_episodes}
+            {
+                "total_steps": total_steps,
+                "total_episodes": total_episodes,
+                "dataset_size": self.get_size(),
+            }
         )
 
 
diff --git a/minari/dataset/_storages/hdf5_storage.py b/minari/dataset/_storages/hdf5_storage.py
index 03559c98..bca359b7 100644
--- a/minari/dataset/_storages/hdf5_storage.py
+++ b/minari/dataset/_storages/hdf5_storage.py
@@ -68,6 +68,8 @@ def update_episode_metadata(
                 ep_group = file[f"episode_{episode_id}"]
                 ep_group.attrs.update(metadata)
 
+        self.update_metadata({"dataset_size": self.get_size()})
+
     def get_episode_metadata(self, episode_indices: Iterable[int]) -> Iterable[Dict]:
         with h5py.File(self._file_path, "r") as file:
             for ep_idx in episode_indices:
@@ -174,7 +176,11 @@ def update_episodes(self, episodes: Iterable[EpisodeBuffer]):
 
         total_steps = self.total_steps + additional_steps
         self.update_metadata(
-            {"total_steps": total_steps, "total_episodes": total_episodes}
+            {
+                "total_steps": total_steps,
+                "total_episodes": total_episodes,
+                "dataset_size": self.get_size(),
+            }
         )
 
 
diff --git a/minari/dataset/minari_storage.py b/minari/dataset/minari_storage.py
index e6743116..deba6929 100644
--- a/minari/dataset/minari_storage.py
+++ b/minari/dataset/minari_storage.py
@@ -339,6 +339,7 @@ def update_from_storage(self, storage: MinariStorage):
             {
                 "author": author1.union(author2),
                 "author_email": email1.union(email2),
+                "dataset_size": self.get_size(),
             }
         )
 
diff --git a/minari/utils.py b/minari/utils.py
index 3e24b8da..0183804a 100644
--- a/minari/utils.py
+++ b/minari/utils.py
@@ -429,8 +429,6 @@ def create_dataset_from_buffers(
 
     storage.update_metadata(metadata)
     storage.update_episodes(buffer)
-    storage.update_metadata({"dataset_size": storage.get_size()})
-
     return MinariDataset(storage)