What “static” and “dynamic” actually mean
Every device that talks on a network needs an IP address. The only question is who decides what that address is, and whether it is allowed to change. That is the whole distinction between static and dynamic.
A dynamic IP is assigned automatically by another machine — usually a DHCP server — and the assignment is temporary. The device asks “what should my address be?” and is told “use this one, for now.” A static IP is configured to a fixed value that does not change on its own: either typed manually into the device, or pinned by a server that always hands back the same address. Nothing about the packets on the wire looks different; the difference is entirely about how the address was chosen and how long it lasts.
The catch that trips people up is that this question applies separately at two levels: the private address a device uses inside your LAN (local network), and the public address your whole household shows to the internet on its WAN (the ISP-facing side). They are governed by different systems and cost different amounts, so keep them strictly apart — more on that below.
How DHCP hands out dynamic addresses
Almost every dynamic address you will encounter is handed out by DHCP, the Dynamic Host Configuration Protocol. When your laptop joins Wi-Fi it broadcasts a request; a DHCP server (your home router, or the ISP's equipment for your public address) answers with an address plus the subnet mask, default gateway, and DNS servers — all in a fraction of a second before you notice you are online.
Leases and renewal
Crucially, DHCP does not give an address away permanently — it grants a lease for a set period (often a day, sometimes an hour or a week). Roughly halfway through, the device quietly asks to renew, and normally the server hands back the same address and extends the timer. This is why a “dynamic” IP so often looks static in practice: as long as a device stays connected and keeps renewing, it usually keeps its address for a long time. The address is most likely to change when a device is off past lease expiry, when the pool is exhausted, or when the network is renumbered.
Addresses come from a defined range called a pool or scope. Because leases free up when devices leave, one small pool can serve far more devices over time than it holds at any instant — which is exactly why dynamic allocation exists.
Why ISPs default to dynamic
ISPs hand out dynamic public addresses by default for the same reason: efficiency. Under IPv4 there are only about 4.3 billion addresses in the entire world, and they ran out years ago. Not every customer is online and actively using an address at the same instant, so leasing addresses dynamically from a shared pool lets an ISP serve more subscribers than it has addresses. A static assignment, by contrast, ties up one scarce address permanently whether or not it is in use.
Dynamic assignment is also easier to operate: the ISP can renumber, rebalance, and do maintenance without coordinating fixed addresses with every customer. That flexibility is worth money to them, which is part of why a guaranteed static public IP is usually sold as a paid add-on. In some networks the scarcity goes a step further and multiple customers share a single public address through carrier-grade NAT — worth understanding before you assume you can simply buy your way to a reachable address.
When you genuinely need a static IP — and when you don't
The honest answer for most home users is: you don't. Everything a typical connection does — browsing, streaming, gaming, video calls, cloud apps — is an outbound connection that you initiate. Outbound connections work fine no matter how often your address changes, because the reply comes straight back to whatever address you are currently using. A changing IP is invisible to you.
A static IP earns its keep when something outside your network needs to reach a fixed address on your network reliably, without you being there to update it. The common cases:
- Self-hosting a public service. A website, game server, or media server that outsiders connect to needs an address that does not move — otherwise every time the IP changes, everyone's bookmarks and DNS records point at the wrong place.
- Remote access to your home or office. Reaching a security camera, NAS, or a VPN server at home from elsewhere is far simpler when the address you dial in to is constant.
- IP allow-listing. Corporate VPNs, database servers, and payment or API gateways often restrict access to a specific list of source IPs. If your address keeps changing, you keep getting locked out; a static IP lets an administrator add you once.
- Running a mail server. Sending email from your own server realistically requires a static IP with a matching reverse-DNS (PTR) record; receivers heavily penalise mail from addresses that look dynamic or residential. This is genuinely hard and usually not worth it, but if you insist on self-hosting mail, a static IP is a hard prerequisite, not a nicety.
Notice that all of these involve inbound connections to a fixed point. If your use case is entirely outbound, you almost certainly do not need to pay for anything.
Static LAN IP vs static public IP — do not confuse them
This is the single most important distinction in the whole topic, and it is where most confusion (and wasted money) comes from.
A static LAN IP: free, and set on your router
Inside your home, devices get private addresses like 192.168.1.x or 10.0.0.x from your router's DHCP server. If you want a specific device — a printer, a NAS, a machine you port-forward to — to always have the same private address, you make a DHCP reservation (sometimes called address reservation or static lease) on the router: you tie that device's MAC address to a chosen address, and the router always hands it back. This is the right way to make a LAN address “static.” It costs nothing, it survives reboots, and it is what port-forwarding rules depend on to keep pointing at the correct machine.
You can instead hard-code a static address directly on the device, but a router reservation is usually cleaner because the router still manages the address and will not accidentally lease the same one to something else.
A static public IP: from your ISP, often paid
Your public address — the single address the internet sees your whole household on, which you can check on the What is my IP page — lives on the WAN side and is controlled entirely by your ISP. No setting on your own router can make it static; only the ISP can assign you a fixed public address, and most sell it as a business feature or a monthly add-on. The key takeaway: pinning a device's LAN address does nothing to stop your public IP from changing. They are two separate layers, and conflating them is why people reserve a LAN address and are baffled when their public IP still rotates.
Alternatives to a static public IP
Before you pay your ISP, know that a static public IP is often the expensive way to solve “reach my home from anywhere.” Cheaper and sometimes better options exist:
- Dynamic DNS (DDNS). A small client on your router or a device watches your public IP and, whenever it changes, updates a hostname (like
myhome.example.net) to point at the new address. Outsiders connect to the name instead of the number, so a changing dynamic IP stops mattering. For most self-hosting hobbyists, free DDNS plus a DHCP reservation is all they ever needed. - Reverse tunnels and relays. Services such as Cloudflare Tunnel, Tailscale, or an SSH reverse tunnel through a cheap VPS let your home device dial out to a relay; visitors reach the relay's public address, which forwards traffic back down the tunnel. Because the connection originates from inside your network, this works even behind CGNAT, where neither a static IP nor port forwarding is available to you.
- A VPN into your network. Running a VPN server at home (or using a mesh VPN) gives you private remote access without exposing services publicly at all. If you are weighing privacy tools generally, the VPN vs proxy vs Tor comparison is a useful primer on what each actually does.
The one case these alternatives cannot fully cover is IP allow-listing by a third party that insists on a fixed source address — there, a genuine static IP (or a static-IP VPN exit) may be unavoidable.
The IPv6 angle
IPv6 changes the framing. Its enormous address space gives every device a globally routable address, so the scarcity that forces IPv4 to be dynamic largely disappears. But IPv6 addresses are not automatically “static”: the network prefix your ISP delegates can still change, and devices deliberately rotate temporary “privacy” addresses for outbound traffic so they are harder to track. So even on IPv6 you may want a stable address for a server — typically a stable interface identifier plus, ideally, a static prefix from the ISP. Run the IPv6 test to check whether your connection has working v6 yet, and see how DNS works for how DDNS and hostnames paper over a changing address regardless of IP version.
Security tradeoffs
A static public IP is slightly more exposed than a dynamic one, though the difference is often overstated. A fixed address is a stable target: it can be scanned, profiled, and added to attackers' lists over time. A changing dynamic IP offers a little accidental obscurity, and means any reputation problems attached to an address — say, if a previous holder was flagged for spam — eventually rotate away from you.
That said, obscurity is not security. Static or dynamic, what actually protects you is not exposing services you did not mean to, keeping them patched, and putting authentication in front of anything reachable. A static IP is neither a meaningful defence nor a serious new risk on its own — reserve your attention for what is actually listening on that address.
Frequently asked questions
Do I actually need a static IP address?
Most people do not. If you only browse, stream, game, and use cloud apps, every connection is outbound and a dynamic IP works perfectly. You typically need a static public IP only when something outside your network must reliably reach a fixed address on it — self-hosting, remote access without a relay, IP allow-listing, or running a mail server. For most of those, a free dynamic DNS hostname is a cheaper substitute than paying your ISP.
What is the difference between a static LAN IP and a static public IP?
A static LAN IP is a fixed private address (like 192.168.1.50) for a device inside your own network, usually set with a DHCP reservation on your router at no cost. A static public IP is a fixed internet-facing address assigned by your ISP — what the rest of the internet sees you on — and ISPs commonly charge for it. Pinning a device's LAN address does nothing to make your public IP stop changing.
Does a dynamic IP change constantly?
Not usually. A dynamic public IP can change, but in practice many stay the same for weeks or months. DHCP hands you an address on a lease and keeps renewing the same one as long as you stay connected and it remains available. It is most likely to change after a long outage, a modem reboot, or an ISP renumbering. If you need it to never change, that is precisely what a static IP or a dynamic DNS hostname is for.
Related reading
- How DHCP works — the protocol that hands out nearly every dynamic address, and where you set reservations.
- What is CGNAT? — why a static IP might not even be an option on some connections.
- Port forwarding explained — the main reason home users want a static LAN address in the first place.
- What is my IP? — check the public address your ISP currently hands your household.
- IPv6 test — see whether your connection has working IPv6, which reframes the whole static-address question.
- How DNS works — how hostnames map to addresses, the basis of dynamic DNS.
- Networking glossary — plain-English definitions for the terms above.