Added growth finder

app.py

@@ -14,6 +14,8 @@ print("IMPORT 4 OK", flush=True)
 from app.ads1.search_term_optimizer import run_search_term_optimizer
 from app.ads1.keyword_inspector import run_keyword_inspector
 from app.ads1.budget_optimizer import run_budget_optimizer
+from app.ads1.growth_finder import run_growth_finder
+from app.ads1.campaign_doctor import run_campaign_doctor
 
 # ==================================================
 # ROMER / ADVISOR DASHBOARD THEME
@@ -1009,6 +1011,9 @@ def campaign_selected(campaign_name, full_state):
     campaign_state = on_campaign_select(full_state, campaign_name)
     return campaign_state, build_campaign_kpi_html(campaign_state.get("campaign_df"))
 
+# def load_campaign_doctor(dfs):
+#     result = run_campaign_doctor(dfs)
+#     return update_right_panel(result)
 
 # ==================================================
 # ADS ANALYST
@@ -1064,6 +1069,23 @@ def run_search_term_optimizer_card(state):
 
     except Exception as e:
         return f"Search term optimization failed: {e}"
+    
+@spaces.GPU(duration=120)
+def run_growth_finder_card(state):
+    try:
+        if not state:
+            return "Select a campaign first."
+
+        dfs = state.get("full_dfs")
+        campaign_name = state.get("campaign_name")
+
+        if dfs is None or campaign_name is None:
+            return "Campaign state is not properly initialized."
+
+        return run_growth_finder(dfs, campaign_name=campaign_name)
+
+    except Exception as e:
+        return f"Search term optimization failed: {e}"
 
 @spaces.GPU(duration=120)
 def run_keyword_inspector_card(state):
@@ -1137,9 +1159,11 @@ with gr.Blocks(fill_height=True, fill_width=True, css=CSS) as demo:
                         value="Search Term Cleaner",
                         elem_classes=["ai-button-card", "search-term-cleaner-card"],
                     )
-                    gr.HTML(ai_card("Growth Finder", "Where to scale?"))
-                    gr.HTML(ai_card("Campaign Doctor", "What limits performance?"))
-
+                    growth_finder_card = gr.Button(
+                        value="Growth Finder",
+                        elem_classes=["ai-button-card", "growth-finder-card"],
+                    )
+                    
                 ads_output = gr.Markdown(
                     value="Select a campaign and click Ads Analyst.",
                     elem_id="ads-output",
@@ -1172,6 +1196,12 @@ with gr.Blocks(fill_height=True, fill_width=True, css=CSS) as demo:
         outputs=ads_output,
     )
 
+    growth_finder_card.click(
+        fn=run_growth_finder_card,
+        inputs=campaign_state,
+        outputs=ads_output,
+    )
+
     keyword_inspector_card.click(
         fn=run_keyword_inspector_card,
         inputs=campaign_state,

app/ads1/growth_finder.py

@@ -0,0 +1,80 @@
+import json
+import pandas as pd
+from app.recs.generate import generate_explanation, is_bad_llm_output
+
+def build_growth_finder_features(df: pd.DataFrame) -> pd.DataFrame:
+    df = df.copy()
+
+    df["cost"] = df["cost"].fillna(0)
+    df["clicks"] = df["clicks"].fillna(0)
+    df["impressions"] = df["impressions"].fillna(0)
+    df["conversions"] = df.get("conversions", 0).fillna(0)
+
+    # Core signals
+    df["ctr"] = (df["clicks"] / df["impressions"].replace(0, 1)) * 100
+    df["cvr"] = (df["conversions"] / df["clicks"].replace(0, 1)) * 100
+    df["cpa"] = df["cost"] / df["conversions"].replace(0, 1)
+
+    # Growth signal (IMPORTANT)
+    df["efficiency_score"] = df["conversions"] / df["cost"].replace(0, 1)
+
+    return df
+
+def build_growth_finder_context(dfs: dict, campaign_name: str | None = None):
+    df = dfs["keywords"].copy()
+
+    if campaign_name and "campaign_name" in df.columns:
+        df = df[df["campaign_name"] == campaign_name]
+
+    df = build_growth_finder_features(df)
+
+    # keep high-signal subset only (not rule-based, just signal control)
+    df = df.sort_values("cost", ascending=False).head(200)
+
+    return {
+        "campaign_name": campaign_name,
+        "keywords": df.to_dict("records")
+    }
+
+def build_growth_finder_prompt(context: dict) -> str:
+    payload = json.dumps(context, indent=2, default=str)
+    name = context.get("campaign_name", "this account")
+
+    return (
+        f"Write 3 to 5 bullet points of actionable Google Ads scaling opportunities for {name}.\n"
+        "Use simple language. One insight per bullet. Start each line with '- '. No intro sentence.\n\n"
+        "Focus on:\n"
+        "- keywords/ad groups that can scale budget profitably\n"
+        "- patterns that show strong conversion efficiency\n"
+        "- underfunded high-performing segments\n"
+        "- expansion opportunities (similar keywords, match types, themes)\n"
+        "- signals of demand that are not fully exploited\n\n"
+        "- Only suggest scaling where performance supports it\n"
+        f"Data (JSON):\n{payload}"
+    )
+
+def run_growth_finder(dfs: dict, campaign_name: str | None = None) -> str:
+    print("\n🚀 [growth_finder] STARTED", flush=True)
+
+    if not dfs or "keywords" not in dfs:
+        return "⚠️ No keyword data available."
+
+    context = build_growth_finder_context(dfs, campaign_name)
+
+    print("🧠 [growth_finder] context built", flush=True)
+
+    prompt = build_growth_finder_prompt(context)
+
+    print("✍️ [growth_finder] prompt built", flush=True)
+
+    result = generate_explanation(prompt)
+
+    if is_bad_llm_output(result):
+        print("⚠️ [growth_finder] fallback triggered", flush=True)
+        return (
+            "- Unable to generate scaling opportunities right now.\n"
+            "- Try again or check data quality."
+        )
+
+    print("📤 [growth_finder] result received", flush=True)
+    return result