The Hidden Half of Demand

Sales data shows what was sold — it hides what customers tried to buy but couldn't. This whitepaper introduces demand reconstruction: quantifying suppressed demand from stockout duration, frequency, and cross-retailer patterns.

Five Key Findings

• Chronically stocked-out SKUs underestimate demand by 18-32% — costing $2.3M+ per $100M in revenue.
• Stockout duration is a stronger demand signal than sales velocity — 3-week backorders indicate 4x more demand than 1-day backorders.
• Cross-retailer stockout correlation reveals category-wide trends — when 3+ retailers stock out simultaneously, it's a demand shift, not a supply issue.
• Search-to-cart abandonment on stock days correlates with next-week search volume — predictive by ~6 days.
• Reconstruction accuracy is ±8% at SKU-week granularity using stockout duration + search + cross-retailer signals combined.

How Demand Reconstruction Works

Demand reconstruction combines four signal types into a single unmet-demand estimate. The model is transparent and auditable — not a black box.

What Demand Reconstruction Actually Returns

Every Product Data Scrape Demand Signals API call returns structured JSON. Here's what a live response looks like:

// Product Data Scrape — Demand Reconstruction API
// GET /v1/demand-recon?sku=YOUR-SKU&window=30d
{
  "sku": "BRAND-SKU-042",
  "reconstruction_period": "2026-03-15 to 2026-04-15",
  "observed_sales_units": 4820,
  "reconstructed_demand": {
    "true_demand_estimate": 6340,
    "suppressed_demand": 1520,
    "suppression_pct": 23.9,
    "confidence_interval": "±8.2%"
  },
  "signal_breakdown": {
    "stockout_duration_days": 12,
    "stockout_frequency": 4,
    "cross_retailer_correlated": true,
    "search_velocity_delta": "+42%"
  },
  "revenue_at_stake": 73620.00,
  "scraped_at": "2026-04-15T14:22:00Z"
}

Request sample dataset → — Free 1,000-row sample delivered within 24 hours, scoped to your target categories and platforms.