diff --git a/README.md b/README.md
index f96ab2c9..61637833 100644
--- a/README.md
+++ b/README.md
@@ -49,8 +49,9 @@
 ## Staging Mail Catcher
 To See mail that has been sent on the staging server you must ssh tunnel into the server
 
-ssh -L 1082:localhost:1080 pulsys@lib-jobs-staging1
-Once the tunnel is open you can see the mail that has been sent on staging [here](localhost:8082)
+    ssh -L 1082:localhost:1080 pulsys@lib-jobs-staging1
+
+Once the tunnel is open you can see the [mail that has been sent on staging here](http://localhost:1082)
 
 ## Getting Started
 
diff --git a/docs/lc_call_slips/testing_on_staging.md b/docs/lc_call_slips/testing_on_staging.md
new file mode 100644
index 00000000..a5b514b6
--- /dev/null
+++ b/docs/lc_call_slips/testing_on_staging.md
@@ -0,0 +1,21 @@
+### Testing LC Call Slips on staging
+
+One convenient way to try your changes on
+production-like data is to run the process on
+staging:
+
+1. Deploy main to staging.
+1. SSH into one of the staging boxes and run `cd /opt/lib-jobs/current && bundle exec rake lib_jobs:process_newly_cataloged_records`
+1. [Open mailcatcher](../../README.md#staging-mail-catcher).
+1. Download one of the CSVs that you'd like to compare.
+1. Repeat the above steps, but with your branch deployed to staging.
+1. Compare the diff between the two CSVs to see
+if it is what you expect.  Note that the order
+of rows may change, and this doesn't indicate
+a problem.
+1. You may also choose to run `bundle exec rake lib_jobs:create_csv_for_selector_comparison`, which
+generates a CSV of all possible records from OCLC,
+including ones that are typically filtered out like
+juvenile materials.  You can use this one to make
+sure that your CSV in mailcatcher includes all the
+expected rows.