The GeoIP User Count database provides insight into the number of expected users
on an IP address and on the associated subnet during the past 24 hours, helping
identify a variety of fraud including unauthorized access to content and click
fraud.

To learn more about the GeoIP User Count database, including terms and pricing,
please
[contact our Enterprise Business team](https://www.maxmind.com/en/solutions/connect-with-a-geoip-expert)
for assistance.

 **Data Privacy**

To comply with data privacy regulations, please monitor our
[Do Not Sell My Personal Information Requests](https://www.maxmind.com/en/accounts/current/do-not-sell-requests)
page (login required) for IP addresses and networks that should not be used for
advertising or marketing purposes. You can automate retrieval of this list using
our [privacy exclusions API](/geoip/privacy-exclusions-api). 


## Binary Database

Binary databases make use of the
[MaxMind DB file format](https://maxmind.github.io/MaxMind-DB/).

You can use the [mmdbinspect tool](https://github.com/maxmind/mmdbinspect) (in
beta), a command line interface built with Go, to look up one or more IPs from
one or more MMDB databases and receive output in a parsable JSON format.

[Learn about the fields in the binary version of this database on our binary
database field reference page.](binary/)

## CSV Database

In addition to our MaxMind DB binary format, we also offer GeoIP and GeoLite
databases in a CSV format suitable for importing into a SQL database. The CSV
files are shipped as a single
[zip file](<https://en.wikipedia.org/wiki/Zip_(file_format)>).




The zip file itself is named `GeoIP2-User-Count-CSV_{YYYYMMDD}.zip`.
The downloaded zip file contains a single directory which in turn
contains several files. That directory is named
`GeoIP2-User-Count-CSV_{YYYYMMDD}`.


The files in this zip archive are:

| Filename | Description |
|----------|-------------|
| `LICENSE.txt` | End user license |
| `COPYRIGHT.txt` | Copyright statement |
| `GeoIP2-User-Count-Blocks-IPv4.csv` | CSV file containing data on IPv4 addresses |
| `GeoIP2-User-Count-Blocks-IPv6.csv` | CSV file containing data on IPv6 addresses |


All the CSV files start with a single header row containing column names. The
specific column names and their contents are detailed below. The files are
encoded as UTF-8.


### Blocks Files

There are two CSV files for network blocks, one each for IPv4 and IPv6 blocks.
These are named `GeoIP2-User-Count-Blocks-IPv4.csv` and
`GeoIP2-User-Count-Blocks-IPv6.csv` respectively.

> **⚠️ Warning**
> 
 Please note that the CSV version of the GeoIP User Count
database contains separate CSV files covering the IPv4 and IPv6 space. All the
fields listed below are present as columns in both CSV files, but the values
associated with IPv6 fields in the IPv4 file will be `null`, and the IPv4 fields
in the IPv6 file will be `null`. 



<div class="table">
  <table>
    <tbody>
      <tr>
        <th>Data field name</th>
        <th>Type</th>
        <th>Data field description</th>
      </tr>

      <tr>
        <td>network</td>
        <td>IP network as a string</td>
        <td>
          This is the IPv4 or IPv6 network in CIDR format such as
          "2.125.160.216/29" or "2001:c50::/32". We offer a utility to convert
          this column to start/end IPs or start/end integers. See
          <a href="#conversion-utility">the conversion utility section</a> for
          details.
        </td>
      </tr>

      <tr>
        <td>ipv4_32</td>
        <td>integer</td>
        <td>
          <p>
            The estimated number of users sharing the IPv4 address during the
            past 24 hours. This is not an exact count of individual users but
            rather a normalized value.
          </p>
          <p>
            <a
              href="https://support.maxmind.com/knowledge-base/articles/maxmind-user-context-data#user-counts"
            >
              Learn more about the normalization and values of the user count on
              our Knowledge Base.
            </a>
          </p>
        </td>
      </tr>

      <tr>
        <td>ipv6_64</td>
        <td>integer</td>
        <td>
          <p>
            The estimated number of users sharing the IPv6 /64 network during
            the past 24 hours. IPv6 /64 networks are usually assigned to
            individual households and can be considered similar from a user
            perspective to an individual IPv4 address. This is not an exact
            count of individual users but rather a normalized value.
          </p>
          <p>
            <a
              href="https://support.maxmind.com/knowledge-base/articles/maxmind-user-context-data#user-counts"
            >
              Learn more about the normalization and values of the user count on
              our Knowledge Base.
            </a>
          </p>
        </td>
      </tr>

      <tr>
        <td>ipv4_24</td>
        <td>integer</td>
        <td>
          <p>
            The estimated number of users sharing the IPv4 /24 network that the
            IP address is contained within during the past 24 hours. This is not
            an exact count of individual users but rather a normalized value.
          </p>
          <p>
            These values should be greater than the values provided in
            <code>ipv4_32</code>
            because they are representative of the number of users across a
            larger number of IP addresses.
          </p>
          <p>
            <a
              href="https://support.maxmind.com/knowledge-base/articles/maxmind-user-context-data#user-counts"
            >
              Learn more about the normalization and values of the user count on
              our Knowledge Base.
            </a>
          </p>
        </td>
      </tr>

      <tr>
        <td>ipv6_48</td>
        <td>integer</td>
        <td>
          <p>
            The estimated number of users sharing the IPv6 /48 network that the
            IP address is contained within during the past 24 hours. This is not
            an exact count of individual users but rather a normalized value.
          </p>
          <p>
            These values should be greater than the values provided in
            <code>ipv6_64</code>
            because they are representative of the number of users across a
            larger number of IP addresses.
          </p>
          <p>
            <a
              href="https://support.maxmind.com/knowledge-base/articles/maxmind-user-context-data#user-counts"
            >
              Learn more about the normalization and values of the user count on
              our Knowledge Base.
            </a>
          </p>
        </td>
      </tr>
    </tbody>
  </table>
</div>


### Conversion Utility

We've created a
[small utility program](https://github.com/maxmind/geoip2-csv-converter) to
allow you to convert a GeoIP CSV file's representation of IP addresses to
another format. You can choose between start/end IP addresses, with the
addresses represented as strings or integers.

The program is available from our
[geoip2-csv-converter GitHub project releases tab.](https://github.com/maxmind/geoip2-csv-converter/releases)


## Example Files

We maintain example files in CSV and MMDB format. The files contain dummy data
rather than real GeoIP data.


### CSV Example Files

We maintain examples of the CSV files as they would be downloaded from the account portal:



- [GeoIP2-User-Count-CSV_Example.zip](/examples/GeoIP2-User-Count-CSV_Example.zip)
  

### MMDB Example Files

We maintain test MMDB files on GitHub:



- [GeoIP2-User-Count-Test.mmdb](https://github.com/maxmind/MaxMind-DB/blob/main/test-data/GeoIP2-User-Count-Test.mmdb)
  

Alternatively, you can [view all of our MMDB test data on GitHub](https://github.com/maxmind/MaxMind-DB/tree/main/test-data).

## Database Sizes

<!-- prettier-ignore-start -->

MaxMind databases can vary in size from release to release. If you are
working with file size limitations that are concerning, you should build
your integrations to fail gracefully in event of a significant size
change.

From May to August 2026, the database files varied in file size and number of networks as
follows:

| Database | CSV File Size | MMDB File Size | IPv4 Networks | IPv6 Networks |
|----------|---------------|----------------|---------------|---------------|
- databaseName: "GeoIP User Count"
  csvSizeRange: "5.60 GB - 6.91 GB"
  mmdbSizeRange: "1.65 GB - 1.99 GB"
  ipv4Range: "146,984,000 - 166,267,000"
  ipv6Range: "57,738,000 - 86,092,000"

The listed file sizes are for unpacked databases. Databases are
downloaded in a compressed format.


<!-- prettier-ignore-end -->

## Database Changes

We may add new data fields to the User Count database at any time.

New database fields are added as new columns to the right of existing columns in our CSV files,
and as additional data in our MMDB files.

Subscribe to our [GeoIP release notes](/geoip/release-notes) to be notified when new
data is added to our databases.

## [GeoIP User Count binary database fields](/geoip/docs/databases/user-count/binary/)
<p>The GeoIP User Count binary database contains the following fields for each
network.</p>


<div class="table">
  <table>
    <tbody>
      <tr>
        <th>Data field name</th>
        <th>Type</th>
        <th>Data field description</th>
      </tr>

      <tr>
        <td>ipv4_24</td>
        <td>uint32</td>
        <td>
          <p>
            The estimated number of users sharing the IPv4 /24 network that the
            IP address is contained within during the past 24 hours. This is not
            an exact count of individual users but rather a normalized value.
          </p>
          <p>
            These values should be greater than the values provided in
            <code>ipv4_32</code> because they are representative of the number
            of users across a larger number of IP addresses.
          </p>