World Energy Data

World Energy Data is the Integrated Oil & Gas catalog's curation play: one analyst's chart-driven reading of the entire world energy system, restructured into country-year rows a warehouse can join. The project is self-funded and single-authored - compiled by Shane White - and it earns its shelf space by doing the consolidation work most teams quietly resent: folding the Energy Institute Statistical Review of World Energy (the series formerly published as BP's Statistical Review), International Energy Agency statistics, the Global Carbon Project, NOAA ESRL and World Bank figures into one coherent narrative, refreshed through the Energy Institute's 2026 Statistical Review.

Five layers hold the material. World Energy Trends charts the global picture. Roughly two dozen National Energy Trends pages profile individual systems - China, USA, India, Russia, Japan, Indonesia, Iran, Canada, Mexico, Brazil, Saudi Arabia, South Korea, Germany, South Africa, Turkiye, Australia, UK, France, UAE, Norway and Singapore, plus Australian export and emissions deep-dives. Essays on the Major Fossil Fueled Nations and on Greenhouse Gas Emissions sit beside an Energy Statistics Guide that explains primary versus final energy and the physical content method of primary-energy equivalency. The two working measures throughout are Total Energy Supply by source and Total Final Consumption.

Within the Datadory catalog - 1,744 datasets averaging 7.81 on quality - this record scores 5/10: unusual interpretive breadth, offset by an inferred field dictionary and thin machine-facing documentation. Get a sample of this dataset and we return rows shaped exactly like the dictionary below, cut to the countries and fuels you name.

The natural key is a triplet - country, year, fuel - with the measure deciding which accounting view the value reports:

# one country-year observation, shaped exactly as delivered
country          : Norway
fuel             : Oil
measure          : Total Energy Supply by source
year             : 2025
value_ej_or_unit : 1.85   (dictionary example)

# the same five columns re-cut for the consumption view
country          : Norway
fuel             : Hydro
measure          : Total Final Consumption
year             : 2025
value_ej_or_unit : pinned to your named fuels when the sample is cut

Read the anatomy rather than the magnitudes. One row hands over geography (Norway), energy source (Oil or any of seven sibling categories), the accounting lens (Total Energy Supply for what a system takes in, Total Final Consumption for what its end users burn), the reference year, and the quantity in exajoules or whatever unit the prefix table declares. Multiply that shape across roughly two dozen national profiles and several decades per country, and a transition study stops being an afternoon of screenshot transcription and becomes a filter.

Five fields define every observation, reconstructed from rendered analyses during the August 2026 research pass and carried at inferred confidence until a live pull pins them. Three identify the observation, one dates it, one quantifies it - there is nothing else hiding in the schema, which is either refreshing or suspicious depending on how many fifteen-column CSVs your team has merged this month.

Geography - the world aggregate first, then roughly two dozen national energy-system profiles: China, USA, India, Russia, Japan, Indonesia, Iran, Canada, Mexico, Brazil, Saudi Arabia, South Korea, Germany, South Africa, Turkiye, Australia, UK, France, UAE, Norway and Singapore, with dedicated Australian export and emissions deep-dives. That mix puts the largest emitters and the major exporters in one frame - rare in a single source at any price, let alone one maintained by a single curator.

Temporal - annual country-year series tracking Energy Institute and IEA publication vintages through the 2026 Statistical Review. Historical depth varies by country because each national series follows its own vintage back through the decades; several profiles run long enough to anchor decade-scale background analysis.

Granularity - country-year by fuel, under either Total Energy Supply or Total Final Consumption. There is no sub-national cut and no company cut anywhere in this dataset; it answers how nations power themselves, not who pumped which barrel.

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

Pick the channel your stack already speaks and set the cadence to match the decision being fed - flat-file loads for research benches, direct pipes into Snowflake, BigQuery or Redshift, lookup calls for anything interactive. An annual series rarely needs hourly plumbing, but if your warehouse prefers one Tuesday load a week, that is a settings conversation, not a contract renegotiation.

Normalization happens before anything reaches you. Chart-rendered aggregates arrive as typed numeric columns with their units resolved against the prefix table, joined to the field dictionary above unchanged. Every shipment includes sample rows for validation and the coverage profile mapped to the countries and fuels you named, so the first thing your pipeline consumes is not the last thing a human squinted at.

Four jobs it does better than any raw statistical portal.

Run cross-country energy comparisons. Market researchers treat it as the pre-reconciled base: roughly two dozen national profiles sharing one accounting spine and one measure vocabulary, ready for the market-sizing and transition-study work that otherwise dies in a swamp of incompatible statistical yearbooks.

Ground journalism and academic work. Citing a curated Energy Institute, IEA and NOAA-based series carries differently in print and peer review than citing a spreadsheet of uncertain provenance - the interpretive essays come with the numbers.

Track competitor exposure between planning cycles. National oil and gas production trends move the competitive picture; a compact annual read of who produces what keeps strategy documents current without a data-engineering project.

Settle the units argument once. The Energy Statistics Guide's treatment of primary versus final energy and the physical content method gives every team a shared vocabulary before the first modeling disagreement, not after the third.

Market Researchers & Consultants score it relevance 3 in the persona pack - see market researchers use cases. Journalists, Academics & Students match that 3 for citable, interpretation-included series - see journalists academics use cases. Competitive Intel & Product Teams carry relevance 2 for exposure tracking - see competitive intel product teams use cases. Strategy and policy analysts round out the bench on the strength of the guide alone. The common thread is a low column count with high interpretive leverage: five fields in, one defensible narrative out.

Within integrated oil and gas data, this record owns curated interpretation: pre-digested national narratives with an energy-accounting education attached. The neighbours own different jobs. The EIA International Energy Statistics cover crude and gas production, reserves and trade for 200+ countries at product and activity detail - far broader machine reach, none of the storytelling. The Eurostat energy statistics harmonise EU balances from 1990 onward for teams whose map ends at the Union's edge. The JODI Oil & Gas World Database brings monthly country-reported flows when annual is a decade too slow.

On the catalog's rubric the split is explicit: this record scores 5/10 against a 7.81 average across 1,744 datasets, a tier shared by 92 records. You trade documented machine interfaces for a finished analytical frame - and for teams writing reports rather than feeds, that trade often lands on the right side.

Four things, stated plainly.

First, the field dictionary is inferred, not published. The five-field schema was reconstructed from rendered analyses rather than documented by the curator; your sample pins every definition against live records before it ships, which is also where column naming locks down for your pipeline.

Second, this is an annual country-year rhythm, not a monitoring feed. Nothing here will tell you what changed last month; pair it with JODI for monthly flows or the Baker Hughes rig count for activity signal, and let this dataset hold the long structural view.

Third, verified content scope is narrower than the full site. Research passes confirmed the world trends material and Norway's national trends directly; the remaining profiles follow the same architecture but were not individually pulled, so tell us which countries matter and we verify those first.

Fourth, reserves context is not a guaranteed column. Reserve discussion appears inside national narratives rather than as a standalone series in review passes; if reserves drive your model, name them in the sample request and we confirm exactly what exists before you build on it.