388 tests · 37 countries · 14 providers

Every IP geolocation provider makes mistakes. See how much.

Measure your own IP See full rankings

Ten seconds, no signup, and your location is never stored.

14 providers, measured continuously against the real device locations of the people who run the test. No self-reported numbers, no synthetic ground truth.

Median error by provider

All four rankings →
  1. BigDataCloud 2.84km 2.63km–3.35km
  2. IPGeolocation 4.84km 3.78km–6.26km
  3. IPInfo 4.87km 3.65km–5.11km
  4. IPRegistry 4.92km 4.45km–5.45km
  5. IPLocate 5.44km 4.18km–8.32km
  6. Pangea IP Intel 5.46km 3.69km–7.23km
Dot is the median error. Bar spans the 95% confidence interval, which is not symmetric. 5 more providers on the full leaderboard.

One Ground Truth. 14 Providers.

Methodology

Three steps. Ten seconds. Real data.

A transparent, reproducible methodology for comparing IP geolocation accuracy against ground truth.

  1. Grant Location Access

    We request device coordinates to establish the reference point. The ground truth all providers are measured against.

  2. Query Every Provider

    Your IP is sent simultaneously to every provider. We capture their city, region, country, and coordinate responses.

  3. Compare Against Truth

    Results are compared against ground truth. Accuracy scores and distance errors are calculated for each provider.

Why

IP geolocation has a measurement problem

Most accuracy claims come from the providers themselves, without any evidence or tested under conditions they control.

  1. Self-reported metrics.

    Providers publish their own accuracy numbers, tested against internal datasets with undisclosed methodology.

    Device location ground truth. Every test uses the contributor's actual location as the reference.

  2. Stale snapshots.

    Academic studies test a fixed set of IPs at a single point in time — they don't reflect how accuracy changes across networks, regions, and ISPs.

    Continuous and live. Rankings update as new tests are contributed. Accuracy trends are tracked over time, not frozen in a paper.

  3. Synthetic ground truth.

    Most benchmarks rely on WHOIS data or known landmarks as reference points — not the actual location of real users on real networks.

    Real networks, real diversity. Tests come from residential broadband, mobile carriers, corporate networks, and hotspots worldwide — the actual conditions providers must serve.

  4. No geographic diversity.

    Testing from a single lab or datacenter captures one slice of network topology, missing mobile carriers, residential ISPs, and emerging markets entirely.

    Fully transparent. Open methodology, open scoring logic, open results. Anyone can verify the results, reproduce the analysis, or contribute tests.

For Developers

Choose with data, not marketing

Stop relying on provider self-assessments. See real accuracy data before you commit to an API integration. Compare city-level precision where it matters most — in the regions you actually serve.

For Providers

Benchmark against the field

Identify weak regions. Understand where your data excels and where it falls short compared to competitors. An independent benchmark gives your accuracy claims credibility that self-reported numbers never can.

For the Industry

Raise the bar for everyone

Transparent, crowdsourced measurement creates competitive pressure to improve. When accuracy is visible, providers invest in better data. The entire ecosystem benefits — better geolocation means better fraud detection, better compliance, and better user experiences.

See how accurate your IP location really is

Ten seconds, no signup, location never stored. You get a personal accuracy report — and every test strengthens the rankings.

Test Your IP Now