Update app.py

app.py

@@ -4,8 +4,6 @@ import pandas as pd
 import plotly.express as px
 import plotly.graph_objects as go
 from plotly.subplots import make_subplots
-import shap
-import matplotlib.pyplot as plt
 from datetime import datetime, timedelta
 from sklearn.model_selection import train_test_split
 from sklearn.compose import ColumnTransformer
@@ -17,7 +15,6 @@ from sklearn.metrics import r2_score, mean_absolute_error
 import warnings
 warnings.filterwarnings('ignore')
 
-# Enhanced page config
 st.set_page_config(
     page_title="Profitability Intelligence Suite",
     page_icon="📊",
@@ -25,7 +22,7 @@ st.set_page_config(
     initial_sidebar_state="collapsed"
 )
 
-# Custom CSS for premium look
+# Custom CSS
 st.markdown("""
 <style>
     .main-header {
@@ -34,7 +31,6 @@ st.markdown("""
         color: #1f77b4;
         text-align: center;
         margin-bottom: 0.5rem;
-        text-shadow: 2px 2px 4px rgba(0,0,0,0.1);
     }
     .sub-header {
         font-size: 1.2rem;
@@ -56,7 +52,6 @@ st.markdown("""
         padding: 1.5rem;
         margin: 1rem 0;
         border-radius: 8px;
-        box-shadow: 0 4px 8px rgba(0,0,0,0.05);
     }
     .recommendation-card {
         background: white;
@@ -65,31 +60,20 @@ st.markdown("""
         padding: 1.5rem;
         margin: 1rem 0;
         box-shadow: 0 4px 12px rgba(0,0,0,0.08);
-        transition: transform 0.2s;
-    }
-    .recommendation-card:hover {
-        transform: translateY(-5px);
-        box-shadow: 0 8px 20px rgba(0,0,0,0.12);
     }
     .positive-impact {
         color: #28a745;
         font-weight: 700;
         font-size: 1.5rem;
     }
-    .stTabs [data-baseweb="tab-list"] {
-        gap: 2rem;
-    }
-    .stTabs [data-baseweb="tab"] {
-        height: 3rem;
-        font-size: 1.1rem;
-        font-weight: 600;
+    .negative-impact {
+        color: #dc3545;
+        font-weight: 700;
+        font-size: 1.5rem;
     }
 </style>
 """, unsafe_allow_html=True)
 
-# -----------------------------
-# Data Generation
-# -----------------------------
 @st.cache_data(show_spinner=False)
 def generate_synthetic_data(days=60, seed=42, rows_per_day=600):
     rng = np.random.default_rng(seed)
@@ -171,67 +155,82 @@ def generate_synthetic_data(days=60, seed=42, rows_per_day=600):
     df = pd.DataFrame(records)
     return df
 
-@st.cache_data(show_spinner=False)
-def build_features(_df):
-    df = _df.copy()
-    # FIXED: Remove duplicate features
-    feats_num = ["net_price", "unit_cost", "qty", "discount_pct", "list_price", "dow"]
-    feats_cat = ["product", "region", "channel"]
-    df = df.sort_values("date").copy()
-    seg = ["product", "region", "channel"]
-
-    # Add rolling features
-    df["roll7_qty"] = df.groupby(seg)["qty"].transform(lambda s: s.rolling(7, min_periods=1).median())
-    df["roll7_price"] = df.groupby(seg)["net_price"].transform(lambda s: s.rolling(7, min_periods=1).median())
-    df["roll7_cost"] = df.groupby(seg)["unit_cost"].transform(lambda s: s.rolling(7, min_periods=1).median())
-
-    feats_num += ["roll7_qty", "roll7_price", "roll7_cost"]
-    target = "gm_pct"
-    return df, feats_num, feats_cat, target
-
-@st.cache_resource(show_spinner=False)
-def train_model(feats_num, feats_cat, target, _X, _y):
-    pre = ColumnTransformer(
-        transformers=[
-            ("cat", OneHotEncoder(handle_unknown="ignore"), feats_cat),
-            ("num", "passthrough", feats_num),
-        ]
-    )
-    model = RandomForestRegressor(n_estimators=80, max_depth=8, random_state=42, n_jobs=-1)
-    pipe = Pipeline([("pre", pre), ("rf", model)])
-    X_train, X_test, y_train, y_test = train_test_split(_X, _y, test_size=0.25, shuffle=False, random_state=42)
-    pipe.fit(X_train, y_train)
-    pred = pipe.predict(X_test)
-    r2 = r2_score(y_test, pred)
-    mae = mean_absolute_error(y_test, pred)
-    return pipe, {"r2": r2, "mae": mae}, X_test
-
-@st.cache_data(show_spinner=False)
-def compute_shap_values(_pipe, _X_sample, feats_num, feats_cat, shap_sample=400):
-    try:
-        np.random.seed(42)
-        X_sample = _X_sample.copy() if hasattr(_X_sample, 'copy') else pd.DataFrame(_X_sample)
-
-        if len(X_sample) > shap_sample:
-            sample_idx = np.random.choice(len(X_sample), size=shap_sample, replace=False)
-            X_sample = X_sample.iloc[sample_idx]
-
-        X_t = _pipe.named_steps["pre"].transform(X_sample)
-        if hasattr(X_t, 'toarray'):
-            X_t = X_t.toarray()
-
-        cat_features = list(_pipe.named_steps["pre"].named_transformers_["cat"].get_feature_names_out(feats_cat))
-        feature_names = cat_features + feats_num
-
-        explainer = shap.TreeExplainer(_pipe.named_steps["rf"])
-        shap_values = explainer.shap_values(X_t)
-
-        shap_df = pd.DataFrame(shap_values, columns=feature_names)
-
-        return shap_df, X_sample.reset_index(drop=True), feature_names
-    except Exception as e:
-        st.error(f"Error computing SHAP: {str(e)}")
-        return None, None, None
+def analyze_margin_bridge(df, current_date, prior_date):
+    """
+    Professional Price-Volume-Mix (PVM) analysis following FP&A best practices
+    Breaks down GM variance into: Price Effect, Volume Effect, Mix Effect, Cost Effect
+    """
+    current_data = df[df["date"] == current_date].copy()
+    prior_data = df[df["date"] == prior_date].copy()
+
+    # Calculate totals for both periods
+    current_total_revenue = current_data["revenue"].sum()
+    current_total_cogs = current_data["cogs"].sum()
+    current_total_gm = current_total_revenue - current_total_cogs
+    current_gm_pct = current_total_gm / current_total_revenue if current_total_revenue > 0 else 0
+
+    prior_total_revenue = prior_data["revenue"].sum()
+    prior_total_cogs = prior_data["cogs"].sum()
+    prior_total_gm = prior_total_revenue - prior_total_cogs
+    prior_gm_pct = prior_total_gm / prior_total_revenue if prior_total_revenue > 0 else 0
+
+    total_gm_variance = current_total_gm - prior_total_gm
+
+    # Aggregate by segment
+    current_seg = current_data.groupby(["product", "region", "channel"]).agg({
+        "revenue": "sum",
+        "cogs": "sum",
+        "qty": "sum",
+        "net_price": "mean",
+        "unit_cost": "mean"
+    }).reset_index()
+    current_seg["gm"] = current_seg["revenue"] - current_seg["cogs"]
+    current_seg["gm_pct"] = current_seg["gm"] / current_seg["revenue"]
+
+    prior_seg = prior_data.groupby(["product", "region", "channel"]).agg({
+        "revenue": "sum",
+        "cogs": "sum",
+        "qty": "sum",
+        "net_price": "mean",
+        "unit_cost": "mean"
+    }).reset_index()
+    prior_seg["gm"] = prior_seg["revenue"] - prior_seg["cogs"]
+    prior_seg["gm_pct"] = prior_seg["gm"] / prior_seg["revenue"]
+
+    # Merge segments
+    merged = pd.merge(
+        current_seg, 
+        prior_seg, 
+        on=["product", "region", "channel"], 
+        suffixes=("_curr", "_prior"),
+        how="outer"
+    ).fillna(0)
+
+    # Price-Volume-Mix Decomposition (industry standard method)
+    # Price Effect: (Current Price - Prior Price) × Current Volume
+    merged["price_effect"] = (merged["net_price_curr"] - merged["net_price_prior"]) * merged["qty_curr"]
+
+    # Volume Effect: (Current Volume - Prior Volume) × Prior Price × Prior GM%
+    merged["volume_effect"] = (merged["qty_curr"] - merged["qty_prior"]) * merged["net_price_prior"] * merged["gm_pct_prior"]
+
+    # Cost Effect: -(Current Cost - Prior Cost) × Current Volume
+    merged["cost_effect"] = -(merged["unit_cost_curr"] - merged["unit_cost_prior"]) * merged["qty_curr"]
+
+    # Mix Effect: Residual (actual GM change minus price/volume/cost effects)
+    merged["gm_variance"] = merged["gm_curr"] - merged["gm_prior"]
+    merged["mix_effect"] = merged["gm_variance"] - (merged["price_effect"] + merged["volume_effect"] + merged["cost_effect"])
+
+    return merged, {
+        "total_gm_variance": total_gm_variance,
+        "price_effect_total": merged["price_effect"].sum(),
+        "volume_effect_total": merged["volume_effect"].sum(),
+        "cost_effect_total": merged["cost_effect"].sum(),
+        "mix_effect_total": merged["mix_effect"].sum(),
+        "current_gm": current_total_gm,
+        "prior_gm": prior_total_gm,
+        "current_gm_pct": current_gm_pct,
+        "prior_gm_pct": prior_gm_pct
+    }
 
 def estimate_segment_elasticity(df, product, region, channel):
     seg_df = df[(df["product"]==product)&(df["region"]==region)&(df["channel"]==channel)]
@@ -285,516 +284,365 @@ def simulate_pricing_action(segment_df, elasticity, discount_reduction_pct):
     except:
         return None
 
-# -----------------------------
 # Main App
-# -----------------------------
-
-st.markdown('<h1 class="main-header">🎯 Profitability Intelligence Suite</h1>', unsafe_allow_html=True)
-st.markdown('<p class="sub-header">AI-Powered Margin Analysis & Strategic Recommendations</p>', unsafe_allow_html=True)
+st.markdown('<h1 class="main-header">🎯 Daily Profitability Variance Analysis</h1>', unsafe_allow_html=True)
+st.markdown('<p class="sub-header">Understanding What Drives Daily Margin Changes</p>', unsafe_allow_html=True)
 
 # Generate data
 with st.spinner("🔄 Loading business data..."):
     df = generate_synthetic_data(days=60, seed=42, rows_per_day=600)
-    df_feat, feats_num, feats_cat, target = build_features(df)
 
-# Calculate KPIs
+# Calculate daily aggregates
 daily = df.groupby("date").agg(
     revenue=("revenue","sum"),
     cogs=("cogs","sum"),
-    gm_value=("gm_value","sum")
+    gm_value=("gm_value","sum"),
+    qty=("qty","sum")
 ).reset_index()
 daily["gm_pct"] = np.where(daily["revenue"]>0, daily["gm_value"]/daily["revenue"], 0.0)
 
-today_row = daily.iloc[-1]
-yesterday_row = daily.iloc[-2] if len(daily) > 1 else today_row
-week_ago_row = daily.iloc[-8] if len(daily) > 7 else today_row
+current_date = daily["date"].max()
+prior_date = current_date - timedelta(days=1)
+current_row = daily[daily["date"]==current_date].iloc[0]
+prior_row = daily[daily["date"]==prior_date].iloc[0]
+week_ago_row = daily.iloc[-8] if len(daily) > 7 else current_row
 roll7 = daily["gm_pct"].tail(7).mean()
 
-# Executive Dashboard Section
-st.markdown("### 📊 Executive Performance Dashboard")
+gm_variance_pp = (current_row["gm_pct"] - prior_row["gm_pct"]) * 100
+gm_variance_dollar = current_row["gm_value"] - prior_row["gm_value"]
+
+# Executive Dashboard
+st.markdown("### 📊 Executive Summary")
 
 col1, col2, col3, col4 = st.columns(4)
 
 with col1:
-    delta_gm = (today_row["gm_pct"] - yesterday_row["gm_pct"]) * 100
     st.metric(
         label="Gross Margin %",
-        value=f"{today_row['gm_pct']*100:.1f}%",
-        delta=f"{delta_gm:+.2f}pp vs yesterday",
+        value=f"{current_row['gm_pct']*100:.2f}%",
+        delta=f"{gm_variance_pp:+.2f}pp",
         delta_color="normal"
     )
 
 with col2:
-    delta_rev = ((today_row["revenue"] - yesterday_row["revenue"]) / yesterday_row["revenue"] * 100) if yesterday_row["revenue"] > 0 else 0
     st.metric(
-        label="Revenue (Today)",
-        value=f"${today_row['revenue']/1e6:.2f}M",
-        delta=f"{delta_rev:+.1f}% DoD",
+        label="Gross Margin $",
+        value=f"${current_row['gm_value']/1e6:.2f}M",
+        delta=f"${gm_variance_dollar/1e6:+.2f}M",
         delta_color="normal"
     )
 
 with col3:
+    revenue_var_pct = ((current_row["revenue"] - prior_row["revenue"]) / prior_row["revenue"] * 100) if prior_row["revenue"] > 0 else 0
     st.metric(
-        label="Gross Margin $ (Today)",
-        value=f"${today_row['gm_value']/1e6:.2f}M",
-        delta=f"${(today_row['gm_value'] - yesterday_row['gm_value'])/1e6:+.2f}M",
+        label="Revenue",
+        value=f"${current_row['revenue']/1e6:.2f}M",
+        delta=f"{revenue_var_pct:+.1f}%",
         delta_color="normal"
     )
 
 with col4:
-    avg_gm_vs_week = (today_row["gm_pct"] - week_ago_row["gm_pct"]) * 100
+    volume_var_pct = ((current_row["qty"] - prior_row["qty"]) / prior_row["qty"] * 100) if prior_row["qty"] > 0 else 0
     st.metric(
-        label="7-Day Avg GM%",
-        value=f"{roll7*100:.1f}%",
-        delta=f"{avg_gm_vs_week:+.2f}pp WoW",
+        label="Volume (Units)",
+        value=f"{current_row['qty']:,.0f}",
+        delta=f"{volume_var_pct:+.1f}%",
         delta_color="normal"
     )
 
-# Trend visualization
-st.markdown("#### 📈 Performance Trend Analysis")
-
-fig_trends = make_subplots(
-    rows=1, cols=2,
-    subplot_titles=("Gross Margin % Trend", "Revenue & Margin $ Trend"),
-    specs=[[{"secondary_y": False}, {"secondary_y": True}]]
-)
-
-fig_trends.add_trace(
-    go.Scatter(
-        x=daily["date"],
-        y=daily["gm_pct"]*100,
-        name="GM%",
-        line=dict(color="#1f77b4", width=3),
-        fill='tozeroy',
-        fillcolor="rgba(31, 119, 180, 0.1)"
-    ),
-    row=1, col=1
-)
-
-fig_trends.add_trace(
-    go.Scatter(
-        x=daily["date"],
-        y=daily["revenue"]/1e6,
-        name="Revenue",
-        line=dict(color="#2ca02c", width=2)
-    ),
-    row=1, col=2
+# Trend chart
+st.markdown("#### 📈 Gross Margin Trend (Last 30 Days)")
+recent_daily = daily.tail(30)
+
+fig_trend = go.Figure()
+fig_trend.add_trace(go.Scatter(
+    x=recent_daily["date"],
+    y=recent_daily["gm_pct"]*100,
+    mode='lines+markers',
+    name="GM%",
+    line=dict(color="#1f77b4", width=3),
+    fill='tozeroy',
+    fillcolor="rgba(31, 119, 180, 0.1)"
+))
+fig_trend.add_hline(y=roll7*100, line_dash="dash", line_color="red", 
+                   annotation_text="7-Day Average", annotation_position="right")
+fig_trend.update_layout(
+    xaxis_title="Date",
+    yaxis_title="Gross Margin %",
+    height=350,
+    hovermode="x unified"
 )
-
-fig_trends.add_trace(
-    go.Scatter(
-        x=daily["date"],
-        y=daily["gm_value"]/1e6,
-        name="GM Value",
-        line=dict(color="#ff7f0e", width=2, dash="dash")
-    ),
-    row=1, col=2, secondary_y=True
-)
-
-fig_trends.update_xaxes(title_text="Date", row=1, col=1)
-fig_trends.update_xaxes(title_text="Date", row=1, col=2)
-fig_trends.update_yaxes(title_text="Gross Margin %", row=1, col=1)
-fig_trends.update_yaxes(title_text="Revenue ($M)", row=1, col=2)
-fig_trends.update_yaxes(title_text="GM Value ($M)", row=1, col=2, secondary_y=True)
-
-fig_trends.update_layout(height=400, showlegend=True, hovermode="x unified")
-st.plotly_chart(fig_trends, use_container_width=True)
+st.plotly_chart(fig_trend, use_container_width=True)
 
 st.markdown("---")
 
-# Train model
-with st.spinner("🤖 Training AI model..."):
-    X = df_feat[feats_num + feats_cat].copy()
-    y = df_feat[target].copy()
-    pipe, metrics, X_test = train_model(feats_num, feats_cat, target, X, y)
-    st.success(f"✅ Model trained: R² = {metrics['r2']:.3f}, MAE = {metrics['mae']:.4f}")
-
-# Compute SHAP once for all tabs
-with st.spinner("🔬 Analyzing profitability drivers..."):
-    shap_df, X_test_sample, feature_names = compute_shap_values(pipe, X_test, feats_num, feats_cat, shap_sample=400)
+# Perform margin bridge analysis
+with st.spinner("🔬 Performing Price-Volume-Mix analysis..."):
+    variance_detail, summary = analyze_margin_bridge(df, current_date, prior_date)
 
-# Tabs for different sections
-tab1, tab2, tab3 = st.tabs(["🔍 Key Drivers Analysis", "🎯 Strategic Recommendations", "🧪 What-If Simulator"])
+# Main Analysis Tabs
+tab1, tab2, tab3 = st.tabs(["📊 Margin Bridge (PVM)", "🔍 Segment Deep Dive", "💡 Pricing Opportunities"])
 
 with tab1:
-    st.markdown("### Understanding What Drives Your Profitability")
-    st.markdown("""
+    st.markdown(f"### Gross Margin Bridge: {prior_date.strftime('%b %d')} → {current_date.strftime('%b %d')}")
+
+    st.markdown(f"""
     <div class="insight-box">
-    <b>🎓 Business Insight:</b> This analysis reveals which business factors and segment combinations have the strongest impact on gross margin.
-    Understanding these drivers helps prioritize strategic initiatives and operational improvements.
+    <b>📋 Variance Summary:</b><br>
+    Gross margin changed by <b>${gm_variance_dollar/1000:+.1f}K</b> ({gm_variance_pp:+.2f} percentage points)<br>
+    from {prior_row['gm_pct']*100:.2f}% to {current_row['gm_pct']*100:.2f}%
     </div>
     """, unsafe_allow_html=True)
 
-    if shap_df is not None and X_test_sample is not None:
-        # Calculate mean absolute SHAP
-        mean_abs = shap_df.abs().mean().sort_values(ascending=False)
-
-        # Map technical names to business names
-        business_name_map = {
-            "discount_pct": "Discount Level",
-            "unit_cost": "Unit Cost",
-            "net_price": "Net Selling Price",
-            "list_price": "List Price",
-            "qty": "Order Quantity",
-            "roll7_qty": "7-Day Avg Quantity",
-            "roll7_price": "7-Day Avg Price",
-            "roll7_cost": "7-Day Avg Cost",
-            "dow": "Day of Week"
-        }
+    # Waterfall Chart - Professional PVM Analysis
+    st.markdown("#### Price-Volume-Mix (PVM) Waterfall Analysis")
+
+    waterfall_data = pd.DataFrame({
+        "Category": [
+            f"{prior_date.strftime('%b %d')}<br>Gross Margin",
+            "Price<br>Effect",
+            "Volume<br>Effect",
+            "Cost<br>Effect",
+            "Mix<br>Effect",
+            f"{current_date.strftime('%b %d')}<br>Gross Margin"
+        ],
+        "Value": [
+            summary["prior_gm"],
+            summary["price_effect_total"],
+            summary["volume_effect_total"],
+            summary["cost_effect_total"],
+            summary["mix_effect_total"],
+            summary["current_gm"]
+        ],
+        "Type": ["absolute", "relative", "relative", "relative", "relative", "total"]
+    })
+
+    fig_waterfall = go.Figure(go.Waterfall(
+        orientation="v",
+        measure=waterfall_data["Type"],
+        x=waterfall_data["Category"],
+        y=waterfall_data["Value"],
+        text=[f"${v/1000:.1f}K" if abs(v) > 100 else f"${v:.0f}" for v in waterfall_data["Value"]],
+        textposition="outside",
+        connector={"line": {"color": "rgb(63, 63, 63)"}},
+        increasing={"marker": {"color": "#28a745"}},
+        decreasing={"marker": {"color": "#dc3545"}},
+        totals={"marker": {"color": "#1f77b4"}}
+    ))
+
+    fig_waterfall.update_layout(
+        title="Gross Margin Variance Breakdown",
+        showlegend=False,
+        height=450,
+        yaxis_title="Gross Margin ($)"
+    )
+    st.plotly_chart(fig_waterfall, use_container_width=True)
 
-        # Get top drivers with business names
-        top_drivers = []
-        for feat, val in mean_abs.head(10).items():
-            bus_name = feat
-            for key, name in business_name_map.items():
-                if key == feat:
-                    bus_name = name
-                    break
-            if feat.startswith("cat__"):
-                parts = feat.replace("cat__", "").replace("product_", "").replace("region_", "").replace("channel_", "")
-                if "product" in feat.lower():
-                    bus_name = f"Product: {parts}"
-                elif "region" in feat.lower():
-                    bus_name = f"Region: {parts}"
-                elif "channel" in feat.lower():
-                    bus_name = f"Channel: {parts}"
-            top_drivers.append({"Driver": bus_name, "Impact Score": val})
-
-        drivers_df = pd.DataFrame(top_drivers)
-
-        col_a, col_b = st.columns([1, 1])
-
-        with col_a:
-            st.markdown("#### Top 10 Profitability Drivers")
-
-            fig_drivers = go.Figure()
-            fig_drivers.add_trace(go.Bar(
-                y=drivers_df["Driver"][::-1],
-                x=drivers_df["Impact Score"][::-1],
-                orientation='h',
-                marker=dict(
-                    color=drivers_df["Impact Score"][::-1],
-                    colorscale='Blues',
-                    line=dict(color='rgb(8,48,107)', width=1.5)
-                ),
-                text=[f"{v:.4f}" for v in drivers_df["Impact Score"][::-1]],
-                textposition='outside',
-            ))
-
-            fig_drivers.update_layout(
-                title="Ranked by Average Impact on Gross Margin",
-                xaxis_title="Impact Score",
-                yaxis_title="",
-                height=500,
-                showlegend=False
-            )
-            st.plotly_chart(fig_drivers, use_container_width=True)
-
-        with col_b:
-            st.markdown("#### Key Insights")
-
-            top_3 = drivers_df.head(3)
+    # Explanation of each component
+    col_exp1, col_exp2 = st.columns(2)
 
-            st.markdown(f"""
-            <div class="insight-box">
-            <b>🥇 Primary Driver:</b> {top_3.iloc[0]['Driver']}<br>
-            <small>Impact Score: {top_3.iloc[0]['Impact Score']:.4f}</small>
-            </div>
-            """, unsafe_allow_html=True)
+    with col_exp1:
+        st.markdown(f"""
+        <div class="insight-box">
+        <b>💰 Price Effect:</b> ${summary['price_effect_total']/1000:+.1f}K<br>
+        <small>Impact of changes in average selling prices across all transactions.
+        Positive = higher prices captured, Negative = price erosion or higher discounts.</small>
+        </div>
+        """, unsafe_allow_html=True)
 
-            st.markdown(f"""
-            <div class="insight-box">
-            <b>🥈 Secondary Driver:</b> {top_3.iloc[1]['Driver']}<br>
-            <small>Impact Score: {top_3.iloc[1]['Impact Score']:.4f}</small>
-            </div>
-            """, unsafe_allow_html=True)
+        st.markdown(f"""
+        <div class="insight-box">
+        <b>📦 Volume Effect:</b> ${summary['volume_effect_total']/1000:+.1f}K<br>
+        <small>Impact of selling more or fewer units at prior period margins.
+        Positive = higher volumes, Negative = volume decline.</small>
+        </div>
+        """, unsafe_allow_html=True)
 
-            st.markdown(f"""
-            <div class="insight-box">
-            <b>🥉 Tertiary Driver:</b> {top_3.iloc[2]['Driver']}<br>
-            <small>Impact Score: {top_3.iloc[2]['Impact Score']:.4f}</small>
-            </div>
-            """, unsafe_allow_html=True)
+    with col_exp2:
+        st.markdown(f"""
+        <div class="insight-box">
+        <b>🏭 Cost Effect:</b> ${summary['cost_effect_total']/1000:+.1f}K<br>
+        <small>Impact of changes in unit costs (COGS).
+        Positive = cost reduction, Negative = cost inflation.</small>
+        </div>
+        """, unsafe_allow_html=True)
 
-            st.markdown("#### Segment Performance Analysis")
-
-            try:
-                cat_cols = ["product", "region", "channel"]
-                joined = pd.concat([X_test_sample[cat_cols].reset_index(drop=True), 
-                                  shap_df.reset_index(drop=True)], axis=1)
-
-                grp = joined.groupby(cat_cols, as_index=False).mean(numeric_only=True)
-                key_shap_cols = [c for c in shap_df.columns if c in grp.columns]
-                grp["net_impact"] = grp[key_shap_cols].sum(axis=1)
-
-                top_negative = grp.nsmallest(8, "net_impact")
-                top_positive = grp.nlargest(8, "net_impact")
-
-                st.markdown("**⚠️ Product-Region-Channel Combinations Reducing Margin:**")
-                for _, row in top_negative.head(5).iterrows():
-                    st.markdown(f"""
-                    <div class="recommendation-card" style="border-left: 4px solid #dc3545; padding: 0.8rem; margin: 0.5rem 0;">
-                    <b>{row['product']}</b> • {row['region']} • {row['channel']}<br>
-                    <small style="color: #dc3545;">Cumulative Impact: {row['net_impact']:.4f}</small>
-                    </div>
-                    """, unsafe_allow_html=True)
-
-                st.markdown("**✅ Product-Region-Channel Combinations Boosting Margin:**")
-                for _, row in top_positive.head(5).iterrows():
-                    st.markdown(f"""
-                    <div class="recommendation-card" style="border-left: 4px solid #28a745; padding: 0.8rem; margin: 0.5rem 0;">
-                    <b>{row['product']}</b> • {row['region']} • {row['channel']}<br>
-                    <small style="color: #28a745;">Cumulative Impact: {row['net_impact']:.4f}</small>
-                    </div>
-                    """, unsafe_allow_html=True)
-
-                # Visualization
-                st.markdown("---")
-                st.markdown("#### Segment Impact Visualization")
-
-                fig_segments = px.treemap(
-                    grp,
-                    path=['product', 'region', 'channel'],
-                    values=grp['net_impact'].abs(),
-                    color='net_impact',
-                    color_continuous_scale='RdYlGn',
-                    title="Product-Region-Channel Combinations Impact on Margin"
-                )
-                fig_segments.update_layout(height=500)
-                st.plotly_chart(fig_segments, use_container_width=True)
-
-            except Exception as e:
-                st.warning(f"Unable to compute detailed segment analysis: {str(e)}")
-    else:
-        st.error("Unable to compute driver analysis. Please check your data.")
+        st.markdown(f"""
+        <div class="insight-box">
+        <b>🔀 Mix Effect:</b> ${summary['mix_effect_total']/1000:+.1f}K<br>
+        <small>Impact of shifts in product, channel, or customer mix.
+        Reflects selling relatively more/less of high-margin items.</small>
+        </div>
+        """, unsafe_allow_html=True)
+
+    # Key Insight
+    dominant_effect = max([
+        ("Price changes", summary['price_effect_total']),
+        ("Volume changes", summary['volume_effect_total']),
+        ("Cost changes", summary['cost_effect_total']),
+        ("Mix shifts", summary['mix_effect_total'])
+    ], key=lambda x: abs(x[1]))
+
+    st.markdown(f"""
+    <div class="{'insight-box' if gm_variance_dollar > 0 else 'warning-box'}">
+    <b>🎯 Key Takeaway:</b><br>
+    The primary driver of today's margin {'improvement' if gm_variance_dollar > 0 else 'decline'} was 
+    <b>{dominant_effect[0]}</b>, contributing ${dominant_effect[1]/1000:+.1f}K to the overall variance.
+    </div>
+    """, unsafe_allow_html=True)
 
 with tab2:
-    st.markdown("### AI-Generated Strategic Recommendations")
+    st.markdown("### Segment-Level Variance Analysis")
     st.markdown("""
     <div class="insight-box">
-    <b>💡 How This Works:</b> The AI identifies segments with margin pressure and suggests specific pricing actions
-    to improve profitability. Recommendations are ranked by expected financial impact.
+    <b>🔍 Detailed Breakdown:</b> Which specific product-region-channel combinations drove the margin change?
     </div>
     """, unsafe_allow_html=True)
 
-    if shap_df is not None and X_test_sample is not None:
-        with st.spinner("🧠 Generating strategic recommendations..."):
-            try:
-                joined = pd.concat([X_test_sample.reset_index(drop=True), shap_df.reset_index(drop=True)], axis=1)
-                joined["key"] = joined["product"] + "|" + joined["region"] + "|" + joined["channel"]
-
-                cand_cols = [c for c in joined.columns if ("discount" in c.lower() or "cost" in c.lower() or "price" in c.lower()) and c in shap_df.columns]
-                seg_scores = joined.groupby("key")[cand_cols].mean().sum(axis=1)
-                worst_keys = seg_scores.sort_values().head(15).index.tolist()
-
-                recs = []
-                for key in worst_keys:
-                    p, r, c = key.split("|")
-                    hist = df[(df["product"]==p)&(df["region"]==r)&(df["channel"]==c)].sort_values("date")
-                    if hist.empty or len(hist) < 50:
-                        continue
-
-                    eps, _ = estimate_segment_elasticity(hist, p, r, c)
-                    prop_disc_pts = np.clip(abs(seg_scores[key])*10, 1.0, 3.0)
-                    sim = simulate_pricing_action(hist, eps, prop_disc_pts)
-
-                    if sim is None or sim["gm_delta_value"] <= 0:
-                        continue
-
-                    daily_transactions = len(hist) / ((hist["date"].max() - hist["date"].min()).days + 1)
-                    annual_impact = sim["gm_delta_value"] * daily_transactions * 365
-
-                    recs.append({
-                        "Segment": f"{p}",
-                        "Region": r,
-                        "Channel": c,
-                        "Current Discount": f"{sim['baseline_discount']:.1f}%",
-                        "Recommended Discount": f"{sim['new_discount']:.1f}%",
-                        "Expected GM Uplift": sim["gm_delta_value"],
-                        "Annual Impact Estimate": annual_impact,
-                        "Current GM%": sim["gm0_pct"]*100,
-                        "Projected GM%": sim["gm1_pct"]*100
-                    })
-
-                recs_df = pd.DataFrame(recs).sort_values("Expected GM Uplift", ascending=False)
-
-                if len(recs_df) > 0:
-                    st.markdown("#### 🏆 Top 3 Priority Actions")
-
-                    for i, (idx, rec) in enumerate(recs_df.head(3).iterrows()):
-                        st.markdown(f"""
-                        <div class="recommendation-card">
-                            <h4>#{i+1}: {rec['Segment']} • {rec['Region']} • {rec['Channel']}</h4>
-                            <p style="font-size: 1.1rem; margin: 0.5rem 0;">
-                                <b>Recommended Action:</b> Reduce discount from <b>{rec['Current Discount']}</b> to <b>{rec['Recommended Discount']}</b>
-                            </p>
-                            <p style="font-size: 1rem; color: #666; margin: 0.5rem 0;">
-                                Current GM: <b>{rec['Current GM%']:.1f}%</b> → Projected GM: <b style="color: #28a745;">{rec['Projected GM%']:.1f}%</b>
-                            </p>
-                            <p class="positive-impact">
-                                💰 Expected Daily Impact: ${rec['Expected GM Uplift']:.2f}
-                            </p>
-                            <p style="font-size: 0.95rem; color: #666;">
-                                📊 Estimated Annual Impact: <b>${rec['Annual Impact Estimate']/1e3:.1f}K</b>
-                            </p>
-                        </div>
-                        """, unsafe_allow_html=True)
-
-                    st.markdown("---")
-                    st.markdown("#### 📋 Complete Recommendations List")
-                    st.dataframe(recs_df, use_container_width=True, height=400)
-
-                    total_daily_impact = recs_df["Expected GM Uplift"].sum()
-                    total_annual_impact = recs_df["Annual Impact Estimate"].sum()
-
-                    st.markdown(f"""
-                    <div class="insight-box" style="background: linear-gradient(135deg, #667eea 0%, #764ba2 100%); color: white; border: none;">
-                        <h3 style="color: white; margin-top: 0;">💎 Total Opportunity</h3>
-                        <p style="font-size: 1.3rem; margin: 0.5rem 0;">
-                            <b>Daily GM Impact:</b> ${total_daily_impact:.2f}
-                        </p>
-                        <p style="font-size: 1.6rem; margin: 0.5rem 0;">
-                            <b>Estimated Annual Impact:</b> ${total_annual_impact/1e6:.2f}M
-                        </p>
-                    </div>
-                    """, unsafe_allow_html=True)
-                else:
-                    st.info("No significant optimization opportunities detected in current data.")
-            except Exception as e:
-                st.error(f"Error generating recommendations: {str(e)}")
-    else:
-        st.error("Unable to generate recommendations. Please check your data.")
+    # Top positive and negative contributors
+    variance_detail_sorted = variance_detail.sort_values("gm_variance", ascending=False)
+
+    col_seg1, col_seg2 = st.columns(2)
+
+    with col_seg1:
+        st.markdown("#### 📈 Top 5 Margin Gainers")
+        top_gainers = variance_detail_sorted.head(5)
+
+        for idx, row in top_gainers.iterrows():
+            if row["gm_variance"] > 0:
+                st.markdown(f"""
+                <div class="recommendation-card" style="border-left: 4px solid #28a745;">
+                <b>{row['product']}</b><br>
+                <small>{row['region']} • {row['channel']}</small><br>
+                <span class="positive-impact">+${row['gm_variance']:.2f}</span><br>
+                <small>
+                • Price Effect: ${row['price_effect']:+.2f}<br>
+                • Volume Effect: ${row['volume_effect']:+.2f}<br>
+                • Cost Effect: ${row['cost_effect']:+.2f}<br>
+                • Mix Effect: ${row['mix_effect']:+.2f}
+                </small>
+                </div>
+                """, unsafe_allow_html=True)
+
+    with col_seg2:
+        st.markdown("#### 📉 Top 5 Margin Losers")
+        top_losers = variance_detail_sorted.tail(5)
+
+        for idx, row in top_losers.iterrows():
+            if row["gm_variance"] < 0:
+                st.markdown(f"""
+                <div class="recommendation-card" style="border-left: 4px solid #dc3545;">
+                <b>{row['product']}</b><br>
+                <small>{row['region']} • {row['channel']}</small><br>
+                <span class="negative-impact">${row['gm_variance']:.2f}</span><br>
+                <small>
+                • Price Effect: ${row['price_effect']:+.2f}<br>
+                • Volume Effect: ${row['volume_effect']:+.2f}<br>
+                • Cost Effect: ${row['cost_effect']:+.2f}<br>
+                • Mix Effect: ${row['mix_effect']:+.2f}
+                </small>
+                </div>
+                """, unsafe_allow_html=True)
+
+    # Detailed table
+    st.markdown("---")
+    st.markdown("#### Complete Segment Variance Table")
+
+    display_variance = variance_detail[[
+        "product", "region", "channel", "gm_variance",
+        "price_effect", "volume_effect", "cost_effect", "mix_effect"
+    ]].sort_values("gm_variance", ascending=False)
+
+    display_variance.columns = [
+        "Product", "Region", "Channel", "GM Variance",
+        "Price Effect", "Volume Effect", "Cost Effect", "Mix Effect"
+    ]
+
+    st.dataframe(display_variance.style.format({
+        "GM Variance": "${:,.2f}",
+        "Price Effect": "${:,.2f}",
+        "Volume Effect": "${:,.2f}",
+        "Cost Effect": "${:,.2f}",
+        "Mix Effect": "${:,.2f}"
+    }).background_gradient(subset=["GM Variance"], cmap="RdYlGn", vmin=-1000, vmax=1000),
+    use_container_width=True, height=400)
 
 with tab3:
-    st.markdown("### Custom What-If Analysis")
+    st.markdown("### Pricing Optimization Opportunities")
     st.markdown("""
     <div class="insight-box">
-    <b>🧪 Interactive Simulation:</b> Test different pricing strategies for specific segments to understand
-    the potential impact on revenue, volume, and profitability.
+    <b>💡 AI Recommendations:</b> Based on segments with declining margins, here are pricing actions to consider.
     </div>
     """, unsafe_allow_html=True)
 
-    last_day = df["date"].max()
-    seg_today = df[df["date"]==last_day][["product","region","channel"]].drop_duplicates()
-
-    col_sim1, col_sim2, col_sim3 = st.columns(3)
-
-    with col_sim1:
-        selected_product = st.selectbox("📦 Select Product", sorted(seg_today["product"].unique()))
-    with col_sim2:
-        selected_region = st.selectbox("🌍 Select Region", sorted(seg_today["region"].unique()))
-    with col_sim3:
-        selected_channel = st.selectbox("🛒 Select Channel", sorted(seg_today["channel"].unique()))
+    # Focus on segments with negative GM variance and negative price effects
+    problem_segments = variance_detail[
+        (variance_detail["gm_variance"] < -50) | 
+        (variance_detail["price_effect"] < -50)
+    ].copy()
+    problem_segments["priority_score"] = problem_segments["gm_variance"]
+    problem_segments = problem_segments.sort_values("priority_score")
+
+    recs = []
+    for _, seg in problem_segments.head(15).iterrows():
+        p, r, c = seg["product"], seg["region"], seg["channel"]
+        hist = df[(df["product"]==p)&(df["region"]==r)&(df["channel"]==c)].sort_values("date")
+
+        if hist.empty or len(hist) < 50:
+            continue
+
+        eps, _ = estimate_segment_elasticity(hist, p, r, c)
+        discount_reduction = 2.0  # Standard 2pp reduction
+        sim = simulate_pricing_action(hist, eps, discount_reduction)
+
+        if sim and sim["gm_delta_value"] > 0:
+            daily_txns = len(hist) / ((hist["date"].max() - hist["date"].min()).days + 1)
+            annual_impact = sim["gm_delta_value"] * daily_txns * 365
+
+            recs.append({
+                "Segment": p,
+                "Region": r,
+                "Channel": c,
+                "Yesterday GM Loss": seg["gm_variance"],
+                "Root Cause": "Price erosion" if seg["price_effect"] < -30 else "Volume decline" if seg["volume_effect"] < -30 else "Cost increase",
+                "Recommended Action": f"Reduce discount from {sim['baseline_discount']:.1f}% to {sim['new_discount']:.1f}%",
+                "Expected Daily GM Uplift": sim["gm_delta_value"],
+                "Estimated Annual Impact": annual_impact
+            })
 
-    seg_hist = df[
-        (df["product"]==selected_product) &
-        (df["region"]==selected_region) &
-        (df["channel"]==selected_channel)
-    ].sort_values("date")
+    recs_df = pd.DataFrame(recs).sort_values("Expected Daily GM Uplift", ascending=False)
 
-    if not seg_hist.empty and len(seg_hist) >= 50:
-        elasticity, _ = estimate_segment_elasticity(seg_hist, selected_product, selected_region, selected_channel)
-        current = seg_hist.iloc[-1]
+    if len(recs_df) > 0:
+        st.markdown("#### 🏆 Top 3 Priority Actions")
 
-        st.markdown(f"""
-        <div class="insight-box">
-        <b>📊 Current State for {selected_product} • {selected_region} • {selected_channel}:</b><br>
-        • Current Discount: <b>{current['discount_pct']*100:.1f}%</b><br>
-        • Net Price: <b>${current['net_price']:.2f}</b><br>
-        • Unit Cost: <b>${current['unit_cost']:.2f}</b><br>
-        • Avg Daily Volume: <b>{seg_hist.tail(7)['qty'].mean():.0f} units</b><br>
-        • Current GM%: <b>{current['gm_pct']*100:.1f}%</b><br>
-        • Price Elasticity: <b>{elasticity:.2f}</b>
-        </div>
-        """, unsafe_allow_html=True)
+        for i, (_, rec) in enumerate(recs_df.head(3).iterrows()):
+            st.markdown(f"""
+            <div class="recommendation-card">
+                <h4>#{i+1}: {rec['Segment']} • {rec['Region']} • {rec['Channel']}</h4>
+                <p><b>Yesterday's Performance:</b> Lost ${abs(rec['Yesterday GM Loss']):.2f} in gross margin</p>
+                <p><b>Root Cause:</b> {rec['Root Cause']}</p>
+                <p><b>Recommended Action:</b> {rec['Recommended Action']}</p>
+                <p class="positive-impact">💰 Expected Daily Recovery: ${rec['Expected Daily GM Uplift']:.2f}</p>
+                <p><small>📊 Annual Impact Estimate: ${rec['Estimated Annual Impact']/1e3:.1f}K</small></p>
+            </div>
+            """, unsafe_allow_html=True)
 
-        st.markdown("#### 🎯 Test Pricing Strategy")
+        st.markdown("---")
+        st.markdown("#### Complete Action Plan")
+        st.dataframe(recs_df, use_container_width=True)
 
-        discount_change = st.slider(
-            "Adjust Discount Level (percentage points)",
-            min_value=-10.0,
-            max_value=5.0,
-            value=0.0,
-            step=0.5,
-            help="Negative values reduce discount (increase price)"
+        st.download_button(
+            label="📥 Download Recommendations (CSV)",
+            data=recs_df.to_csv(index=False).encode("utf-8"),
+            file_name=f"margin_recovery_plan_{current_date.strftime('%Y%m%d')}.csv",
+            mime="text/csv"
         )
-
-        if discount_change != 0:
-            sim = simulate_pricing_action(seg_hist, elasticity, -discount_change)
-
-            if sim:
-                col_res1, col_res2 = st.columns(2)
-
-                with col_res1:
-                    comparison_data = pd.DataFrame({
-                        'Metric': ['Price', 'Volume', 'GM%'],
-                        'Current': [sim['baseline_price'], sim['baseline_qty'], sim['gm0_pct']*100],
-                        'Projected': [sim['new_price'], sim['new_qty'], sim['gm1_pct']*100]
-                    })
-
-                    fig_comp = go.Figure()
-                    fig_comp.add_trace(go.Bar(
-                        name='Current',
-                        x=comparison_data['Metric'],
-                        y=comparison_data['Current'],
-                        marker_color='#94a3b8'
-                    ))
-                    fig_comp.add_trace(go.Bar(
-                        name='Projected',
-                        x=comparison_data['Metric'],
-                        y=comparison_data['Projected'],
-                        marker_color='#3b82f6'
-                    ))
-
-                    fig_comp.update_layout(
-                        title="Current vs. Projected Performance",
-                        barmode='group',
-                        height=350
-                    )
-                    st.plotly_chart(fig_comp, use_container_width=True)
-
-                with col_res2:
-                    st.markdown("#### 📈 Simulation Results")
-
-                    gm_change = sim['gm1_pct'] - sim['gm0_pct']
-
-                    st.metric(
-                        "Gross Margin Impact",
-                        f"{sim['gm1_pct']*100:.1f}%",
-                        f"{gm_change*100:+.1f}pp"
-                    )
-
-                    st.metric(
-                        "Revenue Impact",
-                        f"${sim['new_price'] * sim['new_qty']:.2f}",
-                        f"${sim['revenue_delta']:+.2f}"
-                    )
-
-                    vol_change = sim['new_qty'] - sim['baseline_qty']
-                    st.metric(
-                        "Volume Impact",
-                        f"{sim['new_qty']:.0f} units",
-                        f"{vol_change:+.0f} units"
-                    )
-
-                    st.markdown(f"""
-                    <div class="insight-box" style="margin-top: 1rem;">
-                    <b>💰 Daily P&L Impact:</b><br>
-                    <span style="font-size: 1.5rem; {'color: #28a745' if sim['gm_delta_value'] > 0 else 'color: #dc3545'}">
-                    ${sim['gm_delta_value']:+.2f}
-                    </span>
-                    </div>
-                    """, unsafe_allow_html=True)
-        else:
-            st.info("👆 Adjust the discount slider above to simulate different pricing strategies")
     else:
-        st.warning("⚠️ Insufficient data for selected segment. Please choose a different combination.")
+        st.success("✅ All segments performing well. No immediate pricing interventions needed.")
 
 st.markdown("---")
 st.markdown("""
-<div style="text-align: center; color: #666; padding: 2rem 0;">
-    <small>
-    🔒 Demo Mode: Using synthetic SAP-style data for illustration purposes
-    </small>
+<div style="text-align: center; color: #666; padding: 1rem;">
+    <small>🔒 Demo Mode: Using synthetic SAP-style transaction data for illustration</small>
 </div>
 """, unsafe_allow_html=True)