diff --git a/transform/models/intermediate/performance/int_performance__detector_metrics_agg_daily.sql b/transform/models/intermediate/performance/int_performance__detector_metrics_agg_daily.sql
index 45c08ef4..b551ac5c 100644
--- a/transform/models/intermediate/performance/int_performance__detector_metrics_agg_daily.sql
+++ b/transform/models/intermediate/performance/int_performance__detector_metrics_agg_daily.sql
@@ -1,9 +1,14 @@
-{{ config(materialized='table') }}
+{{ config(
+    materialized="incremental",
+    unique_key=["detector_id", "sample_date"],
+    on_schema_change="sync_all_columns",
+) }}
 
 -- read the station hourly data
 with station_hourly_data as (
     select *
     from {{ ref('int_performance__detector_metrics_agg_hourly') }}
+    where {{ make_model_incremental('sample_date') }}
 ),
 
 -- now aggregate hourly volume, occupancy and speed to daily level
diff --git a/transform/models/marts/performance/performance__station_metric_agg_recent_one_week.sql b/transform/models/marts/performance/performance__station_metric_agg_recent_one_week.sql
index dae6b63b..05acf0d3 100644
--- a/transform/models/marts/performance/performance__station_metric_agg_recent_one_week.sql
+++ b/transform/models/marts/performance/performance__station_metric_agg_recent_one_week.sql
@@ -66,31 +66,30 @@ hourly_station_volume as (
     where sample_date >= dateadd('day', -8, current_date())
 ),
 
+/*
+ * Join station metadata pairs with hourly observations. This could be
+ * expressed more naturally with 2 joins, 1 for ML stations and 1 for HOV,
+ * but doing so incurs a significant performance cost. Instead, we have a
+ * more permissive join and an aggregation.
+ */
 station_with_ml_hov_metrics as (
     select
-        ax.*,
-        ml.sample_date,
-        ml.sample_hour,
-        ml.hourly_volume as ml_hourly_volume,
-        ml.hourly_vmt as ml_hourly_vmt,
-        ml.hourly_vht as ml_hourly_vht,
-        hov.hourly_volume as hov_hourly_volume,
-        hov.hourly_vmt as hov_hourly_vmt,
-        hov.hourly_vht as hov_hourly_vht
-    from closest_station_with_selection as ax
-    inner join hourly_station_volume as ml
+        css.*,
+        hourly.sample_date,
+        hourly.sample_hour,
+        max(case when hourly.station_type = 'ML' then hourly.hourly_volume end) as ml_hourly_volume,
+        max(case when hourly.station_type = 'ML' then hourly.hourly_vmt end) as ml_hourly_vmt,
+        max(case when hourly.station_type = 'ML' then hourly.hourly_vht end) as ml_hourly_vht,
+        max(case when hourly.station_type = 'HV' then hourly.hourly_volume end) as hov_hourly_volume,
+        max(case when hourly.station_type = 'HV' then hourly.hourly_vmt end) as hov_hourly_vmt,
+        max(case when hourly.station_type = 'HV' then hourly.hourly_vht end) as hov_hourly_vht
+    from closest_station_with_selection as css
+    inner join hourly_station_volume as hourly
         on
-            ax.ml_station_id = ml.station_id
-            and ax.direction = ml.direction
-            and ax.active_date = ml.sample_date
-            and ml.station_type = 'ML'
-    inner join hourly_station_volume as hov
-        on
-            ax.hov_station_id = hov.station_id
-            and ax.direction = hov.direction
-            and ml.sample_date = hov.sample_date
-            and ml.sample_hour = hov.sample_hour
-            and hov.station_type = 'HV'
+            (css.ml_station_id = hourly.station_id or css.hov_station_id = hourly.station_id)
+            and css.direction = hourly.direction
+            and css.active_date = hourly.sample_date
+    group by all
 ),
 
 station_metric_agg as (
diff --git a/transform/models/marts/quality/quality__row_count_summary.sql b/transform/models/marts/quality/quality__row_count_summary.sql
index 7c99eef7..3409ebc9 100644
--- a/transform/models/marts/quality/quality__row_count_summary.sql
+++ b/transform/models/marts/quality/quality__row_count_summary.sql
@@ -1,3 +1,6 @@
+{{ config(
+    snowflake_warehouse = get_snowflake_warehouse(size="XL")
+) }}
 with
 
 ML_HV_DETECTOR_STATUS_DAILY_COUNT as (