{
  "id": "dwd_wg5zzqg2",
  "type": "concept",
  "slug": "metadata-map",
  "title": "The metadata map: every piece, one example",
  "summary": "Catalog, glossary, dictionary, lineage, profiling, quality, contracts, semantic layer: what each one is, shown on a single loan-balance column.",
  "status": "draft",
  "classification": "public",
  "authors": [
    {
      "name": "Shan Umasankar",
      "url": "https://dealwithdata.com",
      "role": "author"
    }
  ],
  "assisted_by": [
    "Claude"
  ],
  "scales": [
    "personal",
    "business",
    "enterprise"
  ],
  "tags": [
    "fundamentals",
    "metadata"
  ],
  "terms": [
    "dwd_gubdecks",
    "dwd_6azimhas",
    "dwd_lk4rq6xm",
    "dwd_xzf7arim",
    "dwd_6prph2yy",
    "dwd_zx3rfw43",
    "dwd_2agj6wti",
    "dwd_2xjo4xi3",
    "dwd_gxekubnp",
    "dwd_qywin2zi",
    "dwd_5k2ibrbv",
    "dwd_p6v7pmhk",
    "dwd_qtwdi2p5"
  ],
  "related": [
    "dwd_2p63wxfx",
    "dwd_5hhpphgh",
    "dwd_cb2bgvae"
  ],
  "created": "2026-10-07",
  "updated": "2026-10-07",
  "version": 1,
  "url": "https://dealwithdata.com/c/metadata-map/",
  "markdown_url": "https://dealwithdata.com/c/metadata-map.md",
  "body_markdown": "Governance vocabulary is a pile of overlapping words. The fastest way to\nuntangle it is to hold one piece of data still and look at it through each\nlens in turn.\n\n**The example:** a lending table `LOAN_DTL` with a column `OUTSTANDING_BAL_AMT`.\n\n## Every piece, on one column\n\n| Piece | What it is | On our column |\n|---|---|---|\n| [[technical-metadata]] | Facts about the physical data | `LOAN_DTL.OUTSTANDING_BAL_AMT`, `DECIMAL(18,2)`, Oracle, schema `LND` |\n| [[data-dictionary]] | One system's column-by-column reference | \"Unpaid principal, USD, \u2265 0\" |\n| [[business-glossary]] | The enterprise's words and meanings | *Outstanding Balance*: unpaid principal as of the reporting date |\n| [[business-element]] | Governed data carrying that meaning in context | *Mortgage Outstanding Balance* |\n| [[critical-data-element]] | A flag: this one matters most | Yes, because it feeds regulatory reporting |\n| [[data-catalog]] | Searchable inventory of everything | Search \"loan balance\" \u2192 this column, its owner, its reports |\n| [[data-lineage]] | Where it comes from and goes | Core banking \u2192 batch job \u2192 `LOAN_DTL` \u2192 exposure report |\n| [[data-profiling]] | What the data actually looks like | 12M rows, min 0, max 4.2M, 0.3% null |\n| [[data-quality]] | Rules it must pass | \"\u2265 0 and not null\": 99.7% pass |\n| [[data-contract]] | Producer's promise to consumers | Stays `DECIMAL`, lands by 06:00, nulls under 1% |\n| [[reference-data]] | Shared code lists | Status codes `AC` / `CL` / `CO` on the same table |\n| [[master-data]] | The golden version of an entity | The loan's borrower resolves to Customer #123 |\n| [[semantic-layer]] | Calculations defined once | `Total Exposure = SUM(OUTSTANDING_BAL_AMT)` |\n\n## How they connect\n\n1. **Harvest** the [[technical-metadata]] and the source [[data-dictionary]].\n2. **Map** columns to [[business-element]]s and [[business-term]]s. This is\n   the business \u2194 technical link, and it is the real product.\n3. **Govern** the elements: owners, [[critical-data-element]] flags, policies,\n   [[data-quality]] rules.\n4. **Trace** them with [[data-lineage]].\n5. **Package** them as data products, contracts or a [[semantic-layer]].\n6. **Serve** them to people, systems and AI agents.\n\n## One line each\n\n- **Catalog:** what exists.\n- **Glossary:** what it means.\n- **Dictionary:** what it means *here*.\n- **Lineage:** where it flows.\n- **Profiling:** what it is.\n- **Quality:** whether it's what it should be.\n- **Contract:** what was promised.\n- **Semantic layer:** how to calculate it.\n\n## Three scales\n\n- **Personal:** your spreadsheet of accounts has columns (technical\n  metadata), you know what \"balance\" means (glossary), and you'd notice if a\n  number went negative (data quality). You're already doing this informally.\n- **Business:** a 20-person company's CRM and accounting system disagree on\n  what an \"active customer\" is. A one-page glossary and two quality rules\n  fix more than any tool purchase.\n- **Enterprise:** millions of columns across hundreds of systems. None of\n  this works by hand; the mapping is proposed by machines and approved by\n  people.\n"
}