Updated search optimizer prompt

app/ads1/search_term_optimizer.py

@@ -1,5 +1,6 @@
 import pandas as pd
 from app.recs.generate import TARGET_CPL, generate_explanation, is_bad_llm_output
+import json
 
 def build_search_term_features(df: pd.DataFrame) -> pd.DataFrame:
     df = df.copy()
@@ -35,42 +36,52 @@ def detect_negative_terms(df):
 def detect_review_terms(df):
     return df[
         (df["clicks"] >= 10) &
-        (df["ctr"] < 0.02) &
+        (df["ctr"] < 2) &
         (df["conversions"] == 0)
     ]
 
 def detect_scaling_terms(df):
     return df[(df["conversions"] > 0)].sort_values("cpa", ascending=True)
 
-def classify_search_term(dfs: dict) -> dict:
-    df = dfs["search_terms"].copy()
-    df = build_search_term_features(df)
+# def classify_search_term(dfs: dict) -> dict:
+#     df = dfs["search_terms"].copy()
+#     df = build_search_term_features(df)
 
-    negatives = detect_negative_terms(df)
-    review = detect_review_terms(df)
-    winners = detect_scaling_terms(df)
+#     negatives = detect_negative_terms(df)
+#     review = detect_review_terms(df)
+#     winners = detect_scaling_terms(df)
 
-    return {
-        "negative_keywords": negatives.to_dict("records"),
-        "review_terms": review.to_dict("records"),
-        "winning_terms": winners.head(10).to_dict("records"),
-    }
+#     return {
+#         "negative_keywords": negatives.to_dict("records"),
+#         "review_terms": review.to_dict("records"),
+#         "winning_terms": winners.head(10).to_dict("records"),
+#     }
+
+def build_search_optimizer_prompt(context: dict) -> str:
+    # Convert data to clean JSON string strings for better LLM readability
+    neg_json = json.dumps(context['negative_keywords'], indent=2)
+    rev_json = json.dumps(context['review_terms'], indent=2)
+    win_json = json.dumps(context['winning_terms'], indent=2)
+
+    return f"""You are an expert Google Ads optimization engine.
+        Analyze the following three categories of search terms for the campaign "{context['campaign_name'] or 'All Campaigns'}":
+
+        ### HIGH WASTE TERMS (Add as negatives)
+        {neg_json}
+
+        ### LOW CTR REVIEW TERMS (Needs attention)
+        {rev_json}
+
+        ### WINNING SCALING TERMS (Profitable)
+        {win_json}
 
-def build_search_optimizer_prompt(context):
-    return f"""
-    You are a Google Ads expert.
-    Analyze these search term categories:
-    NEGATIVE:
-    {context['negative_keywords']}
-    REVIEW:
-    {context['review_terms']}
-    WINNERS:
-    {context['winning_terms']}
-    Return:
-    - 3 actions to reduce waste
-    - 2 scaling opportunities
-    - 2 keyword negatives to add immediately
-    """
+        Provide your analysis in a strict JSON format with this exact schema:
+        {{
+            "waste_reduction_actions": ["action 1", "action 2"],
+            "scaling_opportunities": ["opportunity 1", "opportunity 2"],
+            "immediate_negative_keywords": ["keyword 1", "keyword 2"]
+        }}
+        Return ONLY valid JSON. Do not include markdown code blocks, explanations, or extra text."""
 
 def build_search_optimizer_context(dfs: dict, campaign_name: str | None = None):
     df = _search_terms_for_campaign(dfs, campaign_name)