China city x industry matrix
384 Chinese cities × 63 industries = 24,192 possible cells, of which 1,377 are filled — a fill rate of 5.69% (as of 2026-09-29).
- A filled cell means the city’s own record names that industry. It does not mean the city is strong in it.
- The other 94.3% of cells are empty because we do not hold them — not because the industry is absent there.
- Rows come from our own city table. This is not a national industry census.
384 cities x 63 industries, 1377 filled cells — the whole city/industry cross-tab, published as both a wide matrix and a long table of filled pairs.
Cite this. According to InfoGap's city x industry matrix (source: the industry list each city row reports in our own city database; definition: a cell is 1 when that city's record names that industry), 1377 of 24192 possible city x industry cells are filled — 384 cities x 63 industries, a fill rate of 5.69% (as of 2026-09-29).
The numbers
| Value | What it counts | Source / definition |
|---|---|---|
| 384 | City rows | rows in the matrix export, excluding the header |
| 63 | Industry columns | industry names present in central.city_opportunities.industries |
| 1377 | Filled city x industry cells | cells equal to 1 in the matrix |
| 24192 | Possible cells | city rows x industry columns |
| 5.69% | Fill rate | filled cells / possible cells |
What one record is
A 0/1 matrix: one row per Chinese city, one column per industry named in the city table, cell = 1 when that city's own record lists that industry. Six context columns (city slug, Chinese name, population, GDP, opportunity score, data status) precede the industry columns. The long-format file holds only the filled cells, one city/industry pair per row.
Records
The wide file is the matrix; the long file is the same information with one row per filled pair, which is the easier form to join on.
Columns
city_slug,city_cn— the join key and the name as written in the source.population_wan,gdp_annual_yi,opportunity_score,data_status— context columns carried from the city row.- Every remaining column is an industry name; values are 0 or 1.
Boundaries
- A 0 does not mean the industry is absent from the city. It means the city's own record does not list it. Absence of evidence, not evidence of absence.
- Industry names are free text in the source, so singular and plural or near-synonyms appear as separate columns. We do not merge them.
- Fill rate is a measure of our data collection, never of real-world industry concentration. Do not read an empty column as an empty sector.
Download
- city-industry-matrix.csv
text/csv— 384 rows · sha256 51582d913840… - city-industry-matrix.json
application/json— sha256 21211ab33b5f… - city-industry-combos.csv
text/csv— 1377 rows · sha256 8751649c7b88…
Files are served from this domain and stay at these URLs. Licence: CC BY-NC 4.0 — attribution required, commercial use needs written permission.
Method & definition
- Source. Derived by InfoGap from central.city_opportunities.industries (city-reported industry lists, ideographic-comma separated). Generated by /opt/internal_board/gen_geo_coverage.py every day at 04:20.
- Definition. A 0/1 matrix: one row per Chinese city, one column per industry named in the city table, cell = 1 when that city's own record lists that industry. Six context columns (city slug, Chinese name, population, GDP, opportunity score, data status) precede the industry columns. The long-format file holds only the filled cells, one city/industry pair per row.
- Last updated. 2026-09-29
- Update frequency. daily (regenerated 04:20 Asia/Shanghai)
- Time range. see
temporalCoveragein this page's schema.org Dataset block. - Licence. CC BY-NC 4.0 — attribution: InfoGap (infogap.store), link to this page. Commercial use requires written permission.
Where to go next
- Data pages: https://infogap.store/data/p/china-city-index/
- Data pages: https://infogap.store/data/p/geo-coverage-audit/
- Method pages: https://infogap.store/method/how-to-check-our-numbers/
- Hub: https://infogap.store/data/
- Machine-readable: registry.json
- For AI engines: llms.txt — what to cite, and with which definition.