diff --git a/doc/changes/DM-52360.perf.md b/doc/changes/DM-52360.perf.md
new file mode 100644
index 00000000..31976c64
--- /dev/null
+++ b/doc/changes/DM-52360.perf.md
@@ -0,0 +1 @@
+Switch the default `finalJob` implementation to the new `aggregate-graph` command, which makes use of multiple cores much more effectively than `transfer-from-graph`.
diff --git a/doc/lsst.ctrl.bps/quickstart.rst b/doc/lsst.ctrl.bps/quickstart.rst
index b3692bcd..422f0de2 100644
--- a/doc/lsst.ctrl.bps/quickstart.rst
+++ b/doc/lsst.ctrl.bps/quickstart.rst
@@ -1161,7 +1161,7 @@ New YAML Section
      implementation: JOB
      concurrencyLimit: db_limit
      command1: >-
-       ${DAF_BUTLER_DIR}/bin/butler transfer-from-graph
+       ${DAF_BUTLER_DIR}/bin/butler aggregate-graph
        {fileDistributionEndPoint}{qgraphFile}
        {butlerConfig}
        --register-dataset-types
diff --git a/python/lsst/ctrl/bps/etc/bps_defaults.yaml b/python/lsst/ctrl/bps/etc/bps_defaults.yaml
index 1344f439..7b1e39a0 100644
--- a/python/lsst/ctrl/bps/etc/bps_defaults.yaml
+++ b/python/lsst/ctrl/bps/etc/bps_defaults.yaml
@@ -114,6 +114,7 @@ useLazyCommands: True
 executionButler:
   whenCreate: "NEVER"
 
+extraAggregateOptions: ""
 finalJob:
   retryUnlessExit: 2
   updateOutputChain: "--update-output-chain"
@@ -124,12 +125,16 @@ finalJob:
   implementation: JOB
   concurrencyLimit: db_limit
   finalPreCmdOpts: "{defaultPreCmdOpts}"
+  requestCpus: 16
+  requestMemory: 32768
   command1: >-
-    ${DAF_BUTLER_DIR}/bin/butler {finalPreCmdOpts} transfer-from-graph
+    ${DAF_BUTLER_DIR}/bin/butler {finalPreCmdOpts} aggregate-graph
     {fileDistributionEndPoint}{qgraphFile}
     {butlerConfig}
+    -j {requestCpus}
     --register-dataset-types
     {updateOutputChain}
+    {extraAggregateOptions}
 
 # Set a global default memory limit (in MiB) for the automatic memory scaling
 # mechanism. The value, 480 GiB, picked based on the cluster specifications