diff --git a/etl/eligibility/ha_15_32/ha_analysis_batch_3.py b/etl/eligibility/ha_15_32/ha_analysis_batch_3.py
index 28efadd0..3dc4d45f 100644
--- a/etl/eligibility/ha_15_32/ha_analysis_batch_3.py
+++ b/etl/eligibility/ha_15_32/ha_analysis_batch_3.py
@@ -1166,10 +1166,9 @@ def get_epc_data(
             ]
 
             # condition 1 - identified for gbis and not eligible
-            condition_1 = (
-                              identified_for_gbis and not eligibility.gbis_warmfront and not eligibility.eco4_warmfront[
-                              "eligible"]
-                          ) & consider_penultimate_epc
+            condition_1 = (identified_for_gbis and not eligibility.gbis_warmfront
+                           and not eligibility.eco4_warmfront["eligible"]
+                           ) & consider_penultimate_epc
 
             # condition 2 - identified for eco4 and not eligible
             condition_2 = (identified_for_eco4 and not eligibility.eco4_warmfront[
@@ -1246,10 +1245,12 @@ def get_epc_data(
                     "uprn": eligibility.epc["uprn"],
                     "is_estimated": searcher.newest_epc.get("estimated") is not None,
                     "property_type": eligibility.epc["property-type"],
-                    "gbis_eligible": eligibility.gbis_warmfront,
                     "eco4_eligible": eligibility.eco4_warmfront["eligible"],
                     "eco4_message": eligibility.eco4_warmfront["message"],
                     "eco4_strict": eligibility.eco4_warmfront["strict"],
+                    "gbis_eligible": eligibility.gbis_warmfront["eligible"],
+                    "gbis_message": eligibility.gbis_warmfront["message"],
+                    "gbis_strict": eligibility.gbis_warmfront["strict"],
                     "sap": float(eligibility.epc["current-energy-efficiency"]),
                     # Property components
                     "roof": eligibility.roof["clean_description"],
@@ -1279,24 +1280,32 @@ def get_epc_data(
             )
 
             model_api = ModelApi(portfolio_id="-".join([ha_name, "eligibility"]), timestamp=created_at)
+            model_api.MODEL_PREFIXES = ["sap_change_predictions"]
 
-            all_predictions = model_api.predict_all(
-                df=scoring_df,
-                bucket="retrofit-data-dev",
-                prediction_buckets={
-                    "sap_change_predictions": "retrofit-sap-predictions-dev",
-                    "heat_demand_predictions": "retrofit-heat-predictions-dev",
-                    "carbon_change_predictions": "retrofit-carbon-predictions-dev"
-                }
-            )
+            scoring_df["id"] = scoring_df["id"] + "phase=0"
+            # We split up the scoring_df and score
+            predictions = []
+            to_loop_over = range(0, scoring_df.shape[0], 400)
+            for chunk in tqdm(to_loop_over, total=len(to_loop_over)):
+                predictions_dict = model_api.predict_all(
+                    df=scoring_df.iloc[chunk:chunk + 400],
+                    bucket="retrofit-data-dev",
+                    prediction_buckets={
+                        "sap_change_predictions": "retrofit-sap-predictions-dev",
+                    }
+                )
 
-            predictions = all_predictions["sap_change_predictions"].copy()
+                predictions.append(predictions_dict["sap_change_predictions"])
+
+            predictions = pd.concat(predictions)
+            predictions_size = predictions.shape[0]
 
             predictions = predictions.rename(columns={"property_id": "row_id"}).merge(
                 results_df[["row_id", "sap"]], how="left", on="row_id"
             )
+            if predictions.shape[0] != predictions_size:
+                raise ValueError("Predictions size has changed")
             predictions["sap_uplift"] = predictions["predictions"] - predictions["sap"]
-            predictions = predictions.groupby("row_id")["sap_uplift"].sum().reset_index()
 
             results_df = results_df.merge(
                 predictions[["sap_uplift", "row_id"]],
@@ -1339,9 +1348,12 @@ def get_epc_data(
 
             eligibility_assessment = pd.DataFrame(eligibility_assessment)
 
+            # Make sure the results haven't changed in size
             results_df = results_df.merge(
                 eligibility_assessment, how="left", on="row_id"
             )
+            if results_df.shape[0] != len(results):
+                raise ValueError("results has changed size")
 
         # We store the results in S3 as a pickle
         save_pickle_to_s3(
@@ -1809,6 +1821,8 @@ def app():
     loader.load()
     loader.ha_facts_and_figures()
 
+    loader.facts_and_figures.to_csv("facts_and_figures.csv", index=False)
+
     # We load in the additional data required to perform the analysis
     cleaned = read_from_s3(
         s3_file_name="cleaned_epc_data/cleaned.bson",