diff --git a/images/warehousekeeper/warehousekeeper.py b/images/warehousekeeper/warehousekeeper.py
index 2d627e2..1909ad7 100644
--- a/images/warehousekeeper/warehousekeeper.py
+++ b/images/warehousekeeper/warehousekeeper.py
@@ -3,7 +3,7 @@
 
 import boto3
 import click
-from deltalake import DeltaTable
+from deltalake import DeltaTable, WriterProperties
 from loguru import logger
 from pyspark.sql import SparkSession
 
@@ -159,10 +159,32 @@ def vacuum(
 
 
 @cli.command(cls=BaseCommand)
-def optimize(bucket_name: str, database_name_prefix: str):
-    """Run OPTIMIZE against all Delta tables in the given folder"""
+@click.option(
+    "--compression-level",
+    type=click.INT,
+    help="The compression level to use",
+    required=False,
+    default=9,
+)
+@click.option(
+    "--compression-type",
+    type=click.STRING,
+    help="The compression type to use",
+    required=False,
+    default="ZSTD",
+)
+def optimize(
+    bucket_name: str,
+    database_name_prefix: str,
+    compression_type: str,
+    compression_level: int,
+):
+    """Run OPTIMIZE against all Delta tables in the database"""
 
-    spark = spark_builder.getOrCreate()
+    wp = WriterProperties(
+        compression=compression_type,
+        compression_level=compression_level,
+    )
 
     for dt in list_tables(bucket_name=bucket_name, prefix=database_name_prefix):
         logger.info(
@@ -170,9 +192,8 @@ def optimize(bucket_name: str, database_name_prefix: str):
             table=dt.table_uri,
             metadata=dt.metadata(),
         )
-        optimize_query = f"OPTIMIZE delta.`{dt.table_uri}`"
-        logger.info(optimize_query)
-        spark.sql(optimize_query).show(truncate=False)
+        metrics = dt.optimize.compact(writer_properties=wp)
+        logger.info(metrics)
 
 
 @cli.command(cls=BaseCommand)