Data quality
How precise is the geolocation in release 2026-11?
Every geolocation dataset is a mix of precise and coarse answers; few say how much of each. This page does,
for the 14.4M /24s that have a country in release 2026-11. Use the
AccuracyRadius column in the packages to keep only the precision your use needs: for example,
city-level analytics should drop rows over 100 km.
See also the methodology, registered vs used and the country coverage map.
City-precise
43.2%
6,216,511 /24s with an accuracy radius of 100 km or less.
Regional
16.2%
2,328,679 /24s placed within 101-500 km: a region, not a city.
Country level
39.5%
5,688,168 /24s where only the country is known (radius over 500 km).
Sources disagree
5.1%
738,719 /24s where the four source databases (MaxMind GeoLite2, IPinfo Lite, DB-IP Lite, IPLocate) named different countries.
Accuracy radius
Share of the /24s with a country, by published accuracy radius.
Month by month
City-precise share, source disagreement and operator geofeeds in each monthly release.
2026-11sources disagree on 5.1% · 3,041 /24s from operator geofeeds
43.2%city-precise
2026-10sources disagree on 5.6% · 2,685 /24s from operator geofeeds
43.0%city-precise
2026-09sources disagree on 5.3% · 2,344 /24s from operator geofeeds
43.2%city-precise
2026-08sources disagree on 6.2% · 2,478 /24s from operator geofeeds
43.1%city-precise
2026-07sources disagree on 7.1% · 1,592 /24s from operator geofeeds
43.1%city-precise
By regional registry
Shares of each registry's /24s with a country. Abuse contact = a WHOIS abuse e-mail is published for the /24 (46.7% overall); it comes from bulk WHOIS, so ARIN and LACNIC space has almost none in the data.
| RIR | /24 with a country | City-precise | Regional | Country level | Sources disagree | Operator geofeed /24 | Abuse contact |
|---|---|---|---|---|---|---|---|
| ARIN | 6,423,858 | 35.6% | 4.0% | 59.2% | 4.6% | 2,962 | 0.1% |
| APNIC | 3,472,378 | 37.0% | 21.3% | 41.4% | 2.5% | 63 | 100.0% |
| RIPE | 3,344,198 | 58.1% | 33.0% | 6.8% | 9.8% | 16 | 96.9% |
| LACNIC | 710,689 | 65.2% | 13.6% | 21.1% | 0.5% | 0 | 0.0% |
| AFRINIC | 453,747 | 53.3% | 28.4% | 15.6% | 6.4% | 0 | 3.2% |
Largest countries
The 40 countries with the most /24s. The full JSON file covers every country with at least 1,000 /24s.
| Country | /24 | City-precise | Regional | Country level | Sources disagree | Abuse contact |
|---|---|---|---|---|---|---|
| United States | 5,843,324 | 31.1% | 3.9% | 64.6% | 4.2% | 2.3% |
| China | 1,347,571 | 15.7% | 0.6% | 83.5% | 0.7% | 99.9% |
| Japan | 781,854 | 50.7% | 48.5% | 0.1% | 1.4% | 95.2% |
| Germany | 546,493 | 52.8% | 43.5% | 0.1% | 6.8% | 91.2% |
| United Kingdom | 509,535 | 39.8% | 56.8% | 1.4% | 23.5% | 91.1% |
| South Korea | 456,635 | 57.2% | 42.0% | 0.1% | 1.6% | 97.0% |
| Brazil | 335,894 | 67.9% | 6.2% | 24.8% | 2.1% | 1.6% |
| France | 324,719 | 56.2% | 42.1% | 0.2% | 5.2% | 94.1% |
| Canada | 286,172 | 58.5% | 6.0% | 33.7% | 4.5% | 1.2% |
| Italy | 227,073 | 53.7% | 43.8% | 0.1% | 4.1% | 95.7% |
| Netherlands | 214,498 | 96.1% | 0.8% | 0.1% | 8.1% | 88.3% |
| Australia | 209,640 | 56.9% | 8.5% | 32.7% | 3.2% | 87.2% |
| India | 196,465 | 39.8% | 13.2% | 43.9% | 4.2% | 85.8% |
| Russia | 180,041 | 48.4% | 5.5% | 44.9% | 2.8% | 98.9% |
| Taiwan | 147,429 | 60.5% | 37.8% | 0.1% | 2.1% | 96.7% |
| Spain | 139,447 | 68.4% | 30.4% | 0.5% | 3.0% | 95.4% |
| Sweden | 129,664 | 45.3% | 51.0% | 0.1% | 11.6% | 87.2% |
| Mexico | 122,660 | 60.3% | 8.8% | 28.6% | 3.5% | 1.6% |
| South Africa | 117,067 | 36.1% | 60.0% | 0.7% | 6.7% | 10.8% |
| Singapore | 113,039 | 46.3% | 0.1% | 49.5% | 43.0% | 76.1% |
| Switzerland | 98,344 | 96.7% | 0.3% | 0.1% | 12.7% | 88.9% |
| Hong Kong SAR China | 95,296 | 64.6% | 0.5% | 24.8% | 21.8% | 71.4% |
| Egypt | 94,703 | 97.6% | 2.3% | 0.0% | 0.3% | 1.1% |
| Poland | 82,995 | 58.9% | 40.1% | 0.1% | 2.6% | 88.7% |
| Indonesia | 80,877 | 40.8% | 5.2% | 51.5% | 4.3% | 91.5% |
| Argentina | 76,855 | 69.2% | 10.6% | 19.8% | 1.7% | 0.3% |
| Ireland | 74,279 | 70.4% | 23.6% | 0.4% | 21.0% | 51.4% |
| Colombia | 70,351 | 56.9% | 42.7% | 0.1% | 1.1% | 0.4% |
| Türkiye | 67,948 | 53.4% | 45.7% | 0.3% | 2.6% | 95.2% |
| Vietnam | 65,456 | 66.3% | 15.1% | 18.4% | 0.7% | 99.2% |
| Norway | 64,289 | 49.3% | 48.9% | 0.3% | 4.4% | 65.2% |
| Finland | 58,651 | 52.7% | 46.1% | 0.0% | 4.1% | 92.2% |
| Belgium | 56,388 | 98.2% | 0.1% | 0.0% | 13.4% | 86.2% |
| Denmark | 49,815 | 76.3% | 22.4% | 0.0% | 3.4% | 97.7% |
| Saudi Arabia | 48,753 | 67.4% | 27.5% | 0.5% | 7.5% | 95.1% |
| Morocco | 48,125 | 37.0% | 62.8% | 0.0% | 0.4% | 0.3% |
| Austria | 44,998 | 68.4% | 29.9% | 0.0% | 4.4% | 95.0% |
| Iran | 42,794 | 16.4% | 0.3% | 83.1% | 0.9% | 87.3% |
| Chile | 41,662 | 74.3% | 6.1% | 18.1% | 3.5% | 2.0% |
| Thailand | 41,619 | 51.3% | 42.4% | 0.4% | 8.6% | 90.3% |
How to read these figures
- Scope: every /24 with a country (the RIR/IANA unallocated reserve is left out).
- Accuracy radius = the published AccuracyRadius column: an estimated radius in km, set from whether the match resolved to city, region or country precision (1,000 km = country level). City-precise = 100 km or less; regional = 101-500 km; country level = over 500 km.
- Sources disagree = at least two of the four geolocation databases (MaxMind GeoLite2, IPinfo Lite, DB-IP Lite, IPLocate) name different countries for the /24. The published country is then the majority, with ties broken by a fixed source order.
- Operator geofeed = the network operator's own published location (RFC 8805) was used for the city. An operator geofeed overrides the four databases.
- Sources-disagree and operator-geofeed are computed in the build; the packages carry AccuracyRadius only.
- Abuse contact = the /24 has an abuse e-mail in its WHOIS record (the AbuseContact column). It comes from bulk WHOIS, which exists for RIPE NCC, APNIC and AFRINIC space only.