diff --git a/examples/custom-data/README.md b/examples/custom-data/README.md
new file mode 100644
index 0000000..35ccce8
--- /dev/null
+++ b/examples/custom-data/README.md
@@ -0,0 +1,3 @@
+# Pass custom data between stages
+
+Check [this guide](https://www.pipeless.ai/docs/v1/examples/custom-data) to run the example step by step.
diff --git a/examples/custom-data/process.py b/examples/custom-data/process.py
new file mode 100644
index 0000000..17dcd95
--- /dev/null
+++ b/examples/custom-data/process.py
@@ -0,0 +1,19 @@
+def hook(frame_data, _):
+    # Add data to the frame that you can later recover from hooks of subsequent stages. You can also recover it from subsequent hooks of the same stage.
+    # You can use any kind of data.
+    # Integers: frame_data['user_data'] = 100
+    # Floats:   frame_data['user_data'] = 100.5
+    # Strings:  frame_data['user_data'] = "Hello!"
+    # Heterogeneus arrays: frame_data['user_data'] = ["string", 13, 34.6]
+    # Heterogeneus Dictionaries (IMPORTANT: all keys must be strings):
+    frame_data['user_data'] = {
+        "key1": 0,
+        "key2": [1, "3"],
+        "key3": { "inner1": "hola" }
+    }
+
+    # In a later hook you can obtain the data like:
+    # my_data = frame_data['user_data']
+
+    # To connect stages simply give the list to Pipeless when adding a stream:
+    # pipeless add stream --input-uri "file:///home/user/my/path.mp4" --output-uri "screen" --frame-path "stage1,stage2"