{
  "source": "https://archive.ics.uci.edu/dataset/352/online+retail",
  "citation": "Chen, D. (2015). Online Retail. UCI Machine Learning Repository. https://doi.org/10.24432/C5BW33",
  "license": "CC BY 4.0",
  "source_sha256": "43465a06f2ccf7c8b5bd2892bc7defb52f97487934fe93b16ae4c3936424676d",
  "counts": {
    "source_rows": 541909,
    "retained_rows": 525783,
    "non_merchandise_code_rows": 4321,
    "cancelled_invoice_rows": 9288,
    "non_positive_price_rows": 1181,
    "non_positive_quantity_rows": 1336
  },
  "sku_count": 3789,
  "retained_date_range": [
    "2010-12-01",
    "2011-12-09"
  ],
  "totals": {
    "units": "5572048",
    "gross_gbp": "10240335.010000"
  },
  "concentration": {
    "top_1pct_skus_units": {
      "sku_count": 38,
      "share_pct": "18.06301740401374862528104568"
    },
    "top_10pct_skus_units": {
      "sku_count": 379,
      "share_pct": "58.60365883423832673372519404"
    },
    "top_20pct_skus_units": {
      "sku_count": 758,
      "share_pct": "76.10882031166996407783996118"
    },
    "top_50pct_skus_units": {
      "sku_count": 1895,
      "share_pct": "96.17487142967899774014868501"
    },
    "top_1pct_skus_gross_gbp": {
      "sku_count": 38,
      "share_pct": "19.01068322568482063752326400"
    },
    "top_10pct_skus_gross_gbp": {
      "sku_count": 379,
      "share_pct": "61.30491691794758968534956163"
    },
    "top_20pct_skus_gross_gbp": {
      "sku_count": 758,
      "share_pct": "78.29581622252024350519759021"
    },
    "top_50pct_skus_gross_gbp": {
      "sku_count": 1895,
      "share_pct": "95.86191184579224034585563817"
    }
  },
  "skus_with_multiple_positive_unit_prices": 3301,
  "sku_csv_sha256": "a2c1b5a3c6f31cb8d3239a75fb42f54efcc0a482a1f7b54ee3105186b5890816",
  "price_precision": "Unit prices normalized to six GBP decimal places to remove Excel binary serialization artifacts; original workbook retained and hashed. Fractional-penny prices remain possible.",
  "method": "Mutually exclusive exclusions in order: cancelled invoice, non-positive quantity, non-positive price, stock code outside five digits plus up to two capital letters. Positive invoiced units/value, not net realized sales or profit. Exact decimal arithmetic; no customer identifiers retained. Top-N SKU counts round up. Prices can differ for many reasons; no causal discount effect is inferred.",
  "limitations": [
    "2010\u20132011 observations from one UK gifts retailer, including wholesalers; not a current or representative ecommerce benchmark.",
    "No product costs, stock-on-hand, impressions, conversion rates or promotion labels. Cannot infer profit, dead stock, demand response or sell-through.",
    "Cancellations are excluded, not paired against prior invoices. Gross retained invoice value is not net revenue."
  ]
}