IP to Device Count Database
The IP to Device Count dataset provides network-level intelligence on how many devices share an IP address and whether that sharing is concurrent or rotating over time. This goes far beyond simple "residential" or "mobile" labels. The IP to Device Count database reveals whether an IP is used by 1 device, 10, or 10,000+, and whether that sharing happens all at once (concurrent CGNAT) or one device at a time with ISP rotation (dynamic residential).
The database covers currently (as of April 10, 2026) 7.7 million networks globally, with 1.3 million networks exhibiting daily device sharing patterns, and includes validated coverage for major carriers worldwide.
Status: Alpha (Early Access)
Database Schema
The IP to Device Count database contains the following fields:
| Field Name | Example | Data Type | Description |
|---|---|---|---|
network | 103.89.82.50/32 | TEXT | IP network range (CIDR notation) associated with the device count observation. |
asn | AS10214 | TEXT | Autonomous System Number identifying the network owner/operator. |
shared | daily | TEXT | Shortest time window where significant sharing (≥10 devices) is observed: daily, weekly, or monthly. Indicates whether sharing is concurrent or rotating. |
devices_daily | 10 | INTEGER | Observed unique devices per day (logarithmic bucket: 1, 10, 100, 1000, 10000, 100000). Represents order-of-magnitude device count. |
devices_weekly | 10 | INTEGER | Observed devices over 7 days (logarithmic bucket: 1, 10, 100, 1000, 10000, 100000). Used for campaign-duration measurement. |
devices_monthly | 100 | INTEGER | Observed devices over 30 days (logarithmic bucket: 1, 10, 100, 1000, 10000, 100000). Used for list validation and long-term attribution. |
Logarithmic Device Count Buckets
Device counts are classified into 6 logarithmic buckets representing order-of-magnitude ranges. This approach is more statistically robust than false-precise exact counts:
| Devices | Category | Description | Example |
|---|---|---|---|
| 1 | Single device | Individual connection — typically one household device | Home broadband, personal hotspot |
| 10 | Small shared | Small shared network — household, small office, or dynamic IP rotation | Family router, small business |
| 100 | Medium shared | Medium shared network — business, apartment building, or MDU | Enterprise office, apartment complex |
| 1,000 | Large shared | Large shared network — enterprise, university, or cellular CGNAT | University campus, carrier CGNAT |
| 10,000 | Carrier-scale | Regional CGNAT pools, large carrier infrastructure | Spark NZ, NTT Docomo |
| 100,000 | Hyper-scale | Massive concurrent sharing — the largest CGNAT pools globally | TPG Telecom (AU), Bell Canada |
Why logarithmic? The difference between 8,500 and 9,200 devices sharing an IP is less meaningful than the difference between 10 and 100. Order-of-magnitude buckets capture the signal that matters.
Time Windows
The Device Count data measures device sharing across three observation periods to distinguish concurrent sharing from sequential rotation:
| Window | Interpretation | Primary Use Case |
|---|---|---|
| Daily | Devices observed in a 24-hour period | Security blast radius, real-time blocking decisions |
| Weekly | Devices observed over 7 days | Campaign-duration measurement, short-term targeting |
| Monthly | Devices observed over 30 days | List validation, attribution, long-term patterns |
How to Read Across Windows
| Pattern | Daily | Weekly | Monthly | Interpretation |
|---|---|---|---|---|
| Dynamic IP with rotation | 1 | 1 | 10 | Dynamic residential IP. One device at a time, but ISP rotates assignments over the month. Safe for short-window targeting. |
| Concurrent CGNAT sharing | 100+ | 100+ | 100+ | Concurrently shared infrastructure (CGNAT, VPN, university). High blast radius for blocking. |
| Growing rotation | 1 | 10 | 100 | ISP-managed dynamic pool with increasing rotation velocity. |
The shared field indicates the shortest window where ≥10 devices are observed:
daily= concurrent sharing (high coverage radius)weekly= moderate rotationmonthly= slow rotation (likely sequential)
Filename Reference
curl -L https://ipinfo.io/data/device-count.csv.gz?token=$TOKEN -o device_count.csv.gz
curl -L https://ipinfo.io/data/device-count.mmdb?token=$TOKEN -o device_count.mmdb
curl -L https://ipinfo.io/data/device-count.json.gz?token=$TOKEN -o device_count.json.gz
curl -L https://ipinfo.io/data/device-count.parquet?token=$TOKEN -o device_count.parquet
Request Early Access
Device Count is currently in alpha with limited availability. To request access:
- Visit https://ipinfo.io/data/device-count
- Fill out the early access request form
- Specify your primary use case and expected usage volume
Priority is given to companies with active use cases in security, ad tech, data quality, or network operations who can commit to testing and feedback.
Alpha Status
Device Count is currently in alpha release:
- Schema field names are exploratory and may change
- Data refresh cadence is weekly (faster updates planned)
- Coverage is global but may be incomplete in some regions
- Early access is limited to selected customers