What is an IP address?

An IP address (Internet Protocol address) is the unique numeric label that identifies a device on a network. Every phone, laptop, server, and smart bulb that speaks to the internet has one. Its job is simple but essential: it tells data where to go and where it came from, so that a request leaving your browser can find the right server, and the reply can find its way back to you.

The postal analogy

The clearest way to understand an IP address is to think of the postal system. To send a letter, you need a destination address and a return address. The internet works the same way. Every chunk of data — a packet — carries a destination IP address and a source IP address. Routers along the way read the destination, decide the next hop, and pass the packet on, exactly like a chain of sorting offices moving a letter toward its recipient.

This analogy also explains a few things that surprise people. A postal address identifies a place, not a person — anyone living at that address receives the mail. An IP address is the same: it identifies a connection point on the network, not an individual. And just as a large office building shares one street address for hundreds of employees, one IP address often serves a whole household or company.

What an IP address looks like

There are two versions in use today, and they look very different.

IPv4 is the original and still the most common format. It is a 32-bit number, written as four decimal numbers (each from 0 to 255) separated by dots — for example 203.0.113.42. Those four parts are called octets. Because each octet holds 8 bits, the whole address holds 32 bits, which yields about 4.3 billion possible addresses (2³² = 4,294,967,296). That sounded limitless in the 1980s. It is not nearly enough for a planet of billions of people with many devices each.

IPv6 is the modern format, built to solve that shortage. It is a 128-bit number, written as eight groups of four hexadecimal digits separated by colons — for example 2001:0db8:85a3:0000:0000:8a2e:0370:7334. With 128 bits, IPv6 offers roughly 3.4 × 10³⁸ addresses — an almost incomprehensibly large supply, enough to give every grain of sand on Earth its own address many times over. We cover the differences in depth in IPv4 vs IPv6.

Public vs private addresses

Not every IP address is reachable from the open internet, and this distinction trips up a lot of people.

Private addresses are reserved for use inside a local network — your home, an office, a data centre rack. Three IPv4 ranges are set aside for this (defined in RFC 1918):

• 10.0.0.0 – 10.255.255.255
• 172.16.0.0 – 172.31.255.255
• 192.168.0.0 – 192.168.255.255

If your laptop shows an address like 192.168.1.20, that is a private address. It is meaningful only on your local network; the public internet neither sees it nor routes to it. That is why the same private addresses are reused, harmlessly, on millions of separate home networks at once.

Public addresses are globally unique and routable across the internet. Your router holds one public address, assigned by your internet provider, and every device behind it shares that single public identity when reaching the outside world. The translation between your many private addresses and that one public address is done by a technique called NAT (Network Address Translation), described below.

How one address serves a whole house: NAT

When five devices in your home all browse the web at once, the wider internet sees only your router’s single public IP. Network Address Translation is the trick that makes this work. Your router keeps a table mapping each internal conversation to a port number, rewrites outgoing packets to use the public address, and reverses the process for replies — so each device’s traffic finds its way back to the right place.

NAT was originally a workaround for IPv4 scarcity, and it became near-universal. It is also why most home users cannot, by default, run a public server: incoming connections have no entry in the NAT table and get dropped unless you explicitly forward a port. As IPv4 ran short, providers went further still with carrier-grade NAT (CGNAT), sharing a single public IPv4 address among many customers — one reason a public IP often points to a provider’s regional aggregation point rather than to you specifically.

Static vs dynamic addresses

A public address can be assigned in two ways.

A dynamic address is leased to you temporarily and may change — when your router reconnects, when a lease expires, or when the provider reorganises its pool. This is the default for residential connections, and it is handled automatically by a protocol called DHCP (Dynamic Host Configuration Protocol). For everyday browsing, a changing address makes no difference.

A static address stays the same over time. Servers, mail systems, VPN endpoints, and businesses that need a reliable, addressable presence usually pay their provider for one, because services that other machines connect to need a fixed location.

Where addresses come from

IP addresses are not invented at random — they are allocated through a clear hierarchy:

1. The IANA (Internet Assigned Numbers Authority) sits at the top and hands out very large blocks.
2. Five Regional Internet Registries (RIRs) — RIPE NCC (Europe/Middle East), ARIN (North America), APNIC (Asia-Pacific), LACNIC (Latin America), and AFRINIC (Africa) — receive those blocks and allocate smaller ones within their regions.
3. Internet providers and large organisations receive allocations from their RIR.
4. Finally, your provider assigns an individual address to your connection, and your router assigns private addresses to your devices.

This is the same registry system that records which autonomous system (network operator) announces which blocks — the data behind ipdex itself. An IP address is only useful because that chain of records exists.

From an IP to the organisation behind it

An IP address on its own is just a number. Its real value is that it can be resolved to the organisation that operates the network it belongs to — its ASN (autonomous system number), the operator’s name, the country of registration, and more. This is the bridge from a raw address to a real-world entity.

For example, the public DNS resolver 8.8.8.8 belongs to Google’s network, AS15169 (Google LLC) — a real operator you can reach from a single real address. The live card at the end of this article shows that data directly from the ipdex index, with a link to the full profile. (If a particular address is not in the index, the lookup returns an honest “not found” rather than a guess — accuracy over invention.)

What an IP address can — and cannot — tell you

Because an IP maps to a registered network, you can learn a fair amount from one: the provider or hosting company, the country and often the city or region, whether it belongs to a residential ISP, a mobile carrier, a cloud platform, or a corporate network, and whether it is a known VPN or proxy exit.

What an IP address cannot tell you is just as important. It does not contain your name. It does not reveal your exact home address — geolocation is an estimate, frequently accurate only to the city and sometimes pointing to your provider’s hub many kilometres away. It does not expose your files, your messages, or your browsing history. And because addresses are shared (through NAT and CGNAT) and reassigned (through DHCP), an IP is a weak identifier for a person — which is exactly why serious tracking relies on other signals.

Myth-busting

Myth: “My IP address is my personal ID and can identify me by name.” No. An IP identifies a network connection point, not a person. It is routinely shared by everyone behind a router and reassigned over time. Linking an address to a named individual at a specific moment generally requires records held privately by the provider.

Myth: “If someone has my IP, they can hack into my computer.” Knowing an address is not the same as having access. Your router drops unsolicited incoming connections by default, and your IP is shared with every website you visit as a normal part of how the internet works.

Myth: “An IP address pinpoints my house on a map.” IP geolocation is an approximation derived from registration and routing data, not GPS. It is often correct only to the city, and CGNAT can place you at a provider aggregation point far from your actual location.

Myth: “We are completely out of IP addresses.” The free pool of new IPv4 addresses was exhausted — IANA’s central pool in 2011, with the regional registries running down over the following decade. But IPv4 still works everywhere through reuse, NAT, and a transfer market, and IPv6 provides an effectively unlimited supply going forward.

Myth: “Changing my IP address makes me anonymous.” A new IP changes one signal among many. Modern tracking also uses cookies, logins, and browser fingerprinting, none of which a fresh IP address resets.

Key takeaways

• An IP address is a unique numeric label that tells internet data where to go and where it came from.
• IPv4 (32-bit, ~4.3 billion addresses) is the original format; IPv6 (128-bit) was created to remove the shortage.
• Private addresses work only on local networks; one shared public address represents a whole household to the outside world, via NAT.
• Addresses are leased dynamically (DHCP) or fixed (static), and allocated through IANA → the five RIRs → providers → you.
• An IP’s real value is the chain of records that links it to a real network operator — which is exactly what ipdex indexes.