If you've used the internet, you've used an IP address — billions of times, probably. Every page you load, every video you stream, every push notification that lights up your phone involves dozens of IP addresses quietly shuffling packets back and forth. And yet, for most people, "IP address" is just a vaguely technical phrase that shows up on the occasional support page.

This article is a friendly tour of what an IP address actually is. By the end, you'll understand what the numbers mean, why there are two flavors (IPv4 and IPv6), and how the global system manages to find one specific machine out of the roughly thirty billion devices currently online.

The mailing address analogy

The cleanest way to think about an IP address is as a postal address. When you send a letter, you don't physically carry it to the recipient — you hand it to the postal system, which uses the address to figure out where it should go. The address has structure: country, state, city, street, house number. Each level narrows the search.

An IP address works the same way. When your laptop sends a request to load a webpage, that request is wrapped in a packet that includes a destination IP address. The packet is then handed to your local network, which hands it to your ISP, which hands it to the wider internet. At each step, a router looks at the address and decides which direction to send it next. None of these routers know the entire internet — they only need to know "given this destination, which neighbor do I forward it to?" That's enough to get the packet across the planet.

Reading an IPv4 address

The classic IP address you've probably seen looks like this:

192.168.1.42

This is an IPv4 address. It's made of four numbers separated by dots, each between 0 and 255. Under the hood, this is just a 32-bit number — 32 zeros and ones — written in a way humans can stomach. The dotted-decimal notation breaks the 32 bits into four chunks of 8 bits (a byte), and writes each chunk as a number from 0 to 255.

So 192.168.1.42 is actually:

11000000.10101000.00000001.00101010

This matters because the structure of the address — the way the bits are split between "network part" and "host part" — is how routers know which direction to send a packet. The first portion identifies a network (or sub-network), and the rest identifies a specific host within that network. The dividing line isn't fixed; it varies based on something called a subnet mask, which we'll skip for now.

Public vs private addresses

Some IP ranges are reserved for private networks — addresses that are only meaningful inside your home or office, not on the public internet. The most common ones are:

  • 10.0.0.0 to 10.255.255.255
  • 172.16.0.0 to 172.31.255.255
  • 192.168.0.0 to 192.168.255.255

If your home router gave your laptop the address 192.168.1.42, that address is unique on your network but completely unremarkable globally — millions of other home networks use the same range. The trick that makes this work is called NAT (Network Address Translation): your router holds a single public IP address that the outside world sees, and it rewrites packets coming and going so that all the devices on your private network can share that one public address.

When you visit our "What is my IP" tool, the address it shows is your public IP — the one your router presents to the world. The private address your laptop holds internally is invisible to websites.

Where IPv6 fits in

IPv4 has a fundamental limit: 32 bits gives you about 4.3 billion unique addresses. That sounded like a lot in 1981. It's nowhere near enough today. The official pool of unallocated IPv4 addresses ran dry in 2011, and the regional registries that manage what's left have been rationing them ever since.

IPv6 is the long-term answer. It uses 128-bit addresses, which gives you 340 undecillion of them — a number so large that "we can never run out" is, for once, a reasonable claim. An IPv6 address looks like this:

2001:0db8:85a3:0000:0000:8a2e:0370:7334

It's eight groups of four hexadecimal digits separated by colons. There are shorthand rules — leading zeros in each group can be dropped, and one run of all-zero groups can be replaced with :: — so the address above is more commonly written as:

2001:db8:85a3::8a2e:370:7334

Same address, less typing.

Most modern operating systems and ISPs support IPv6, but the global rollout has been slow because IPv4 still works fine for everyone who doesn't need new public addresses. We have a whole article on the IPv4-vs-IPv6 transition if you want the longer story.

How does my computer get an IP?

You almost never type your own IP address. Instead, when your device joins a network, it asks the network for one. The protocol that handles this is called DHCP (Dynamic Host Configuration Protocol). Your router runs a DHCP server, your device asks "may I have an IP, please?", and the router hands one out from a pool, along with extra info like which DNS server to use and where the default gateway is. The address you get is leased for a period of time, after which the device renews the lease or gets a new one.

For the public IP your router holds, the same thing happens one level up: your router asks your ISP, and the ISP hands one out. Some ISPs give you a "static" IP that doesn't change; most give you a "dynamic" one that may rotate every time you reboot the router or after some interval.

From IP to name (and back)

Humans aren't great at remembering numbers. We are decent at remembering words. That's why the internet has a parallel naming system called DNS (the Domain Name System). When you type example.com into your browser, your computer first asks a DNS server "what's the IP for example.com?", gets back something like 93.184.215.14, and then makes the actual connection to that IP.

DNS also works in reverse: given an IP, you can sometimes ask "what hostname is associated with this address?" — that's a reverse DNS or PTR record lookup. Reverse DNS is best-effort: only the operator of the network can configure it, so not every IP has a useful PTR record. When one is configured, it usually reveals something about the owner. An IP belonging to a residential ISP might reverse to something like ip-203-0-113-42.example-isp.net, while a corporate mail server might reverse to mail.bigcompany.com.

What can someone learn from your IP?

A common worry, especially in privacy discussions, is "what does an IP reveal about me?". The honest answer: less than people fear, but more than zero.

From a public IP, anyone (using a tool like ours) can typically determine:

  • The country and region the IP is registered in
  • The general metro area — often correct, sometimes wildly off
  • The ISP or hosting provider that owns the address
  • The ASN (autonomous system number) of the network operator
  • Whether the IP looks like a datacenter, mobile carrier, or known VPN exit

What they cannot determine from the IP alone:

  • Your name
  • Your home address or phone number
  • What you're doing online
  • Your browsing history

Only your ISP knows which specific customer was using a given IP at a given time, and that information is generally only disclosed in response to a legal request. We dig into the precision question in our geolocation guide.

The TL;DR

  • An IP address is a number that identifies a network endpoint, the same way a mailing address identifies a building.
  • IPv4 addresses are 32-bit and look like 8.8.8.8; IPv6 addresses are 128-bit and look like 2001:db8::1.
  • Your devices each have a private IP on your home network, but they share one public IP that the outside world sees.
  • DNS translates human-readable names like example.com to IP addresses (and reverse DNS does the opposite).
  • A public IP reveals roughly where you are and who your ISP is — not who you are personally.
Want to try it? Look up your own IP with the What is my IP tool, or look up any IP address (like 8.8.8.8) with the IP Lookup tool.

Next up: IPv4 vs IPv6: why the world is slowly moving on — the longer story of why the transition has dragged on for two decades.