diff --git a/test_data/preprocess.sh b/test_data/preprocess.sh
new file mode 100755
index 0000000..ab4dbeb
--- /dev/null
+++ b/test_data/preprocess.sh
@@ -0,0 +1,11 @@
+#!/bin/bash
+
+collections=`find . | grep "\\.completions$"`
+
+for collection in $collections; do
+    echo $collection
+    python extract_dict.py $collection
+    python map_dataset.py $collection
+    python build_stats.py $collection.mapped
+    python build_inverted_and_forward.py $collection
+done