diff --git a/dvc.lock b/dvc.lock
index 6ec804ba..d0b093f1 100755
--- a/dvc.lock
+++ b/dvc.lock
@@ -33,20 +33,20 @@ stages:
     outs:
     - path: input/assessment_data.parquet
       hash: md5
-      md5: 5187eb62e2a44b99247356211a5b02d7
-      size: 308810165
+      md5: 28c8299ff7945f0e1ccfe854339435ce
+      size: 309900279
     - path: input/char_data.parquet
       hash: md5
-      md5: 70a5d4c4d80559462adf09d8cd113c00
-      size: 612547374
+      md5: 00ee5862a32841db25bdd04ee101e708
+      size: 616501379
     - path: input/complex_id_data.parquet
       hash: md5
-      md5: f977b7d16738fa9033fde627fc3feaf5
-      size: 702763
+      md5: 33556a9ed684d14edb3adc17bf109dae
+      size: 702489
     - path: input/hie_data.parquet
       hash: md5
-      md5: da793d132f70a14f917635a434cf027e
-      size: 1924350
+      md5: 8d7ad3974f98db6beb6f93eebe4ea092
+      size: 1928855
     - path: input/land_nbhd_rate_data.parquet
       hash: md5
       md5: e508daf5790982c303d6503fe1cb8e2b
@@ -56,8 +56,8 @@ stages:
       size: 2109
     - path: input/training_data.parquet
       hash: md5
-      md5: f90e28c15c0a71edaa4d7ec0257ebc16
-      size: 156744321
+      md5: 00170e54ba27d99cfb343b9cbe2c0288
+      size: 156769394
   train:
     cmd: Rscript pipeline/01-train.R
     deps:
diff --git a/pipeline/00-ingest.R b/pipeline/00-ingest.R
index bf35512a..351b2833 100644
--- a/pipeline/00-ingest.R
+++ b/pipeline/00-ingest.R
@@ -63,6 +63,25 @@ training_data <- dbGetQuery(
 )
 tictoc::toc()
 
+# NOTE: This is a temporary shim to insert updated sales validation flags
+# that use slightly different groupings/methods vs the production flags. This
+# will be replaced once the production flags are stable
+sales_flags <- read_parquet(paste0(
+  "s3://ccao-ci-test-township-partition-data-warehouse-us-east-1",
+  "/sale/flag/2024-01-22_11:55-charming-damon.parquet"
+))
+
+# Replace the old flags with the new flags by reference
+library(data.table)
+conflicts_prefer(dplyr::between)
+conflict_prefer_all("lubridate")
+setDT(training_data)
+setDT(sales_flags)
+
+training_data[sales_flags, c("sv_is_outlier", "sv_outlier_type") := {
+  .(i.sv_is_outlier, i.sv_outlier_type)
+}, on = .(meta_sale_document_num)]
+
 # Pull all ADDCHARS/HIE data. These are Home Improvement Exemptions (HIEs)
 # stored in the legacy (AS/400) data system
 tictoc::tic("HIE data pulled")