OIV Databases Hub — GI/AO & Vine Varieties

It is the vocabulary layer of the wine world, kept by the one body entitled to keep it. The International Organisation of Vine and Wine — the intergovernmental organisation whose member states set vitivinicultural practice — maintains two structured registries beside its statistics. The GI/AO database records geographical indications and appellations of origin notified to the organisation by those member states. The companion registry is the global catalogue of vine varieties, indexing the varieties used in wine, table grape and dried grape production.

The distinction inside the GI/AO layer is doing legal work, not cosmetic work. A geographical indication says a product comes from a place with a given quality, reputation or character; an appellation of origin goes further and ties that quality or character to the place and the practices carried out within it. The registry stores the class as each member notified it, alongside the product scope the protection covers — wine, table grapes, dried grapes, spirit drinks of vitivinicultural origin. That combination turns a label phrase into a testable claim: when a bottle says protected origin, there is a row behind it, with a member state attached.

One row per registered indication, four typed values apiece — the record shape as it lands in an extract:

registry      : gi_ao
designation   : Rioja                country : Spain
category      : appellation_of_origin
product_scope : wine

designation   : Champagne            country : France
category      : appellation_of_origin
product_scope : wine

designation   : Vale dos Vinhedos    country : Brazil
category      : geographical_indication
product_scope : wine

Read a row top to bottom and you have the whole legal posture of a wine region compressed into four fields: the name as notified, the member standing behind it, whether it claims the stricter appellation class, and what the protection actually covers. The third row is the interesting one — a New World indication classified as a geographical indication rather than an appellation, which is exactly the kind of distinction that breaks naive string matching if your pipeline treats all protected origins alike. The examples above illustrate the record shape; Get a sample of this dataset and the rows come back live, cut to the countries you care about.

Four typed fields ride on every GI/AO record, fully documented in the dictionary below. One honesty flag before you build on them: these definitions were reconstructed from observed registry behaviour rather than transcribed from a published data dictionary, so they describe what arrives in a Datadory extract — and the sample request is where any residual ambiguity dies.

Additional fields on request: the vine-variety catalogue's own columns (variety name, use grouping across wine, table grape and dried grape production, synonym spellings), multilingual forms of designation names across the six languages the organisation publishes in, and notification dates where the registry holds them. Tell us the join and we confirm availability for your scope before delivery.

• Geography — OIV member states and observers worldwide. Because the registry follows notifications, Old World appellations and New World indications land in one table under the same four fields, which is precisely what cross-market comparison needs.
• Time — registry snapshots. A delivery reflects the notifications on file when you ask; new designations enter as member states notify them rather than on a calendar. There is no built-in history, so teams tracking movement retain successive deliveries and diff them.
• Grain — one record per registered indication or per vine variety. No pre-aggregated counts, no regional rollups: everything downstream is an aggregation you control from the base grain.

No published row total exists for either registry, so coverage is quoted as breadth — members reached, classes covered, product scopes protected — rather than a frozen count. Your sample doubles as the census: cut it to your countries of interest and the returned rows are the definitive figure for your scope.

API, files, or your warehouse. Daily, weekly, or hourly.

You pick the channel and the cadence; normalisation is our problem. The registries publish their content as interactive visual reports rather than documented flat files — which is exactly why this exists as a maintained feed instead of a one-time download. Our pipeline resolves the rendered layer into the typed rows above, and that layout stays fixed in every channel: same field names, same types, same join keys whether the records arrive as files, stream through the API, or write straight into Snowflake, BigQuery or Redshift. Extended columns you ask for arrive alongside the canonical four, typed and documented the same way. A sample ships first either way, sized to test inside your pipelines the same day.

• Origin-claim validation. Cross-check the regional wording in listings, menus and marketing against registered designations before it ships — the category and product_scope fields tell you not just whether a name is protected but what the protection covers.
• Competitive portfolio mapping. Group a rival's range by appellation and permitted variety to see which regions their portfolio can legitimately claim, and which claims outrun the registry.
• Controlled vocabularies for wine analytics. Variety-name drift is the silent killer of pooled wine corpora; a stable catalogue keyed to the intergovernmental spelling keeps groupings intact across sources.
• Market-entry research. Segment categories by appellation system — which members lean on indications versus full appellations, and at what product scope — to read regulatory posture as market structure.
• Citation-grade verification. Reporters and academics anchor protected-origin statements to a registry maintained by the organisation itself rather than to a trade association's word.

Market researchers and consultants get denominators and segmentation axes: indication counts by member, class splits and product scopes that anchor category sizing. Journalists, academics and students get verification — a named registry behind every protected-origin sentence they write or teach. Data scientists and ML engineers get controlled vocabularies for designation and variety names; pipeline guidance sits on the data scientists hub. Developers and builders get a lookup table purpose-built for origin-validation features; integration context lives on the developers & builders hub. Competitive intelligence and product teams get the claim map — who can say what, where. E-commerce operators get listing insurance: a check that runs before the SKU goes live, not after the complaint does.

Cards worth reading next: the sibling OIV statistics service that answers how much where this hub answers what may be called where, and the wine-review and physicochemical corpora that give the vocabulary something to attach to. The head-to-head comparisons sort benchmark data against trade-flow alternatives for the wider slice.