---

name: commerce-analytics description: Use when analyzing sales performance, customer metrics, inventory health, or generating forecasts.

Commerce Analytics Skill

Domain knowledge for business intelligence, sales analytics, customer insights, and demand forecasting.

Analytics Concepts

Time Periods

Key Metrics

Sales Summary Structure

Copy{
  totalRevenue: 45230.00,      // Sum of all order totals
  orderCount: 156,             // Number of orders
  averageOrderValue: 290.00,   // Revenue / Orders
  itemsSold: 423,              // Sum of quantities
  uniqueCustomers: 89          // Distinct customers
}

Interpreting Sales Data

• Revenue up, Orders down = Higher AOV (upselling working)
• Revenue down, Orders up = Lower AOV (discounting impact)
• New customers high = Marketing working
• Returning customers high = Retention strong

Product Performance

Top Products Output

Copy{
  sku: "WIDGET-001",
  name: "Premium Widget",
  unitsSold: 78,
  revenue: 15600.00,
  orderCount: 45
}

Product Metrics to Watch

Product Performance Ratios

CopyRevenue per Unit = Revenue / Units Sold
Units per Order = Units Sold / Order Count

Customer Analytics

Customer Metrics Output

Copy{
  totalCustomers: 1250,           // All customers
  newCustomers: 45,               // New in period
  returningCustomers: 234,        // Repeat buyers
  averageLifetimeValue: 450.00,   // Average total spend
  averageOrdersPerCustomer: 2.3   // Order frequency
}

Customer Segmentation

Top Customers Output

Copy{
  customerId: "uuid",
  name: "John Doe",
  email: "john@example.com",
  orderCount: 15,
  totalSpent: 2340.00,
  averageOrderValue: 156.00
}

Inventory Health

Health Overview Output

Copy{
  totalSkus: 150,           // All products
  inStockSkus: 120,         // Available > 0
  lowStockSkus: 18,         // Below reorder point
  outOfStockSkus: 12,       // Available = 0
  totalValue: 125000.00     // Sum of inventory value
}

Stock Status Levels

Low Stock Item Output

Copy{
  sku: "WIDGET-001",
  name: "Premium Widget",
  onHand: 15,
  allocated: 5,
  available: 10,
  reorderPoint: 20,
  averageDailySales: 3.5,
  daysOfStock: 2.8    // available / averageDailySales
}

Demand Forecasting

Forecast Output

Copy{
  sku: "WIDGET-001",
  name: "Premium Widget",
  averageDailyDemand: 3.5,     // Historical average
  forecastedDemand: 105,       // Next 30 days
  confidence: 0.7,             // 70% confidence
  currentStock: 45,
  daysUntilStockout: 12,       // stock / daily demand
  recommendedReorderQty: 105,  // 30 days supply
  trend: "Rising"              // Rising, Falling, Stable
}

Demand Trends

Reorder Recommendations

When daysUntilStockout < leadTime:

• Order immediately
• Quantity = forecastedDemand + safetyStock

Revenue Forecasting

Revenue Forecast Output

Copy{
  period: "Period +1",          // Future period
  forecastedRevenue: 48000.00,  // Point estimate
  lowerBound: 40800.00,         // 80% confidence lower
  upperBound: 55200.00,         // 80% confidence upper
  confidenceLevel: 0.8,         // 80%
  basedOnPeriods: 12            // Historical periods used
}

Forecast Granularity

Interpreting Confidence Intervals

CopyForecasted: $48,000
Lower (80%): $40,800
Upper (80%): $55,200

There's an 80% chance revenue will fall between
$40,800 and $55,200

Order Status Breakdown

Status Breakdown Output

Copy{
  pending: 12,      // Awaiting confirmation
  confirmed: 8,     // Confirmed, not processing
  processing: 15,   // Being prepared
  shipped: 45,      // In transit
  delivered: 120,   // Completed
  cancelled: 5,     // Cancelled by customer/merchant
  refunded: 3       // Refunded
}

Operational Health Indicators

Return Metrics

Return Metrics Output

Copy{
  totalReturns: 23,
  returnRatePercent: 4.5,
  totalRefunded: 3450.00
}

Return Rate Benchmarks

Common Return Reasons

1. Defective - Quality control issue
2. Wrong item - Fulfillment error
3. Not as described - Listing accuracy
4. Changed mind - Normal behavior
5. Better price found - Competitive pricing
6. Damaged - Shipping issue

Analytics Workflows

Weekly Business Review

Copy1. get_sales_summary(period: "last7days")
2. get_top_products(period: "last7days", limit: 5)
3. get_order_status_breakdown(period: "last7days")
4. get_return_metrics(period: "last7days")

Monthly Planning

Copy1. get_sales_summary(period: "last_month")
2. get_customer_metrics(period: "last_month")
3. get_revenue_forecast(periodsAhead: 3, granularity: "month")
4. get_demand_forecast(daysAhead: 30)

Inventory Review

Copy1. get_inventory_health()
2. get_low_stock_items(threshold: 20)
3. get_demand_forecast(skus: [critical_skus], daysAhead: 14)

Best Practices

1. Compare periods - Always show context vs prior period
2. Focus on trends - Single data points are less meaningful
3. Segment analysis - Break down by product, customer, region
4. Action-oriented - Every insight should suggest an action
5. Set benchmarks - Define what "good" looks like for your business
6. Regular cadence - Schedule analytics reviews weekly/monthly

Common Questions and Answers

"Is this a good month?"

Compare to:

• Same month last year (seasonality)
• Previous month (trend)
• Monthly average (benchmark)

"Which products should I reorder?"

Look for:

• daysUntilStockout < 14
• trend = "Rising"
• High revenue contribution

"Are customers coming back?"

Check:

• returningCustomers / (totalCustomers - newCustomers)
• Average orders per customer
• Customer lifetime value trend

"Is my pricing right?"

Analyze:

• AOV trends
• Return rate
• Revenue per unit sold