When a business plans for growth, the conversation usually centres on traffic, conversions, and customers. Almost nobody brings up IP addresses. They sit quietly underneath every website, app, and checkout page, and most of the time your hosting or cloud provider takes care of them.
But IPv4 addresses are in short supply, and that shortage has real effects. It shows up in cloud bills, in what your provider can offer you, and in how smoothly an expansion goes. You don’t need to become a network engineer to deal with it. Knowing the basics helps you ask your suppliers better questions, and helps you tell a genuine infrastructure need from money you don’t need to spend.
Why IPv4 is running out
Every device and service on the internet needs an address so traffic can find it. IPv4, the protocol most of the internet still relies on, uses 32-bit addresses. That gives roughly 4.3 billion combinations, and a chunk of those are reserved for private networks and other special uses.
That sounded like plenty when the protocol was designed. It isn’t anymore. Techniques that let many devices share one address have stretched the supply, but they can’t add a single new address to the pool.
The regional registries have been out for years. ARIN, which covers Canada, the United States, and parts of the Caribbean, gave out the last of its general free pool on 24 September 2015. The RIPE NCC, which serves Europe, the Middle East, and parts of Central Asia, made its final allocation on 25 November 2019.
Running out doesn’t mean anything stopped working. Existing addresses carry on as normal. What’s gone is the large supply of never-used addresses that registries used to hand out. If you need more today, they come from addresses already in circulation, supplied through a provider, bought through a transfer, or leased.
This isn’t just an IT problem
Most online shops, agencies, and small businesses will never buy an address block. Their hosting platform or cloud provider handles it.
The effects still reach them, mostly through pricing and technical limits. If you run a store on a managed ecommerce platform, you may never notice. If you’re a SaaS company running your own cloud environment, you’re more likely to see IPv4 charges on your invoice and to hit caps on how many addresses you can allocate.
It’s also worth clearing up a common assumption: more customers don’t automatically mean more IP addresses. Thousands of visitors can reach the same website through shared infrastructure. What drives address needs is how your systems are built, not how much traffic you get.
So for most growing businesses, IPv4 scarcity is something to plan around rather than panic about. It matters most when growth means new servers, separate environments for individual clients, or launching in new regions.
The cost shows up on your bill
Cloud pricing makes this concrete.
Since February 2024, AWS has charged US$0.005 per hour for every public IPv4 address, whether it’s actively in use or sitting idle. Across a standard 730-hour month, that’s about US$3.65 per address. A hundred addresses come to roughly US$365 a month, before any credits or taxes. (Addresses a business already owns and brings to AWS through its BYOIP programme aren’t subject to this charge.)
Keep in mind this is one provider’s fee for using an address. It isn’t the market price for buying or leasing IPv4 space, and it doesn’t include compute, storage, bandwidth, or anything else on the bill.
For a small setup, a few dollars a month is easy to ignore. It adds up once you have production, staging, and test environments spread across several regions. Addresses left behind after a project ends keep getting billed too.
A periodic review of address usage, alongside the rest of your cloud spend, usually pays for itself. Releasing addresses nobody needs can cut costs without touching anything customers see. Just have your technical team confirm nothing still depends on them first.
Plan for addresses before you launch
New markets and products often bring infrastructure needs that don’t come up in commercial planning.
A regional launch might need extra application servers, network gateways, or integrations built for specific customers. Some business clients also insist on fixed IP addresses so they can allow your connections through their firewalls.
Sort these out before you set a launch date. Find out whether your provider can supply what you need, what restrictions apply, and whether the setup will hold up as you grow. Some questions worth asking:
- Which services actually need a public IPv4 address?
- Can several applications safely share one entry point?
- Do any customer integrations need an address that never changes?
- Does your cloud account or region have allocation limits?
- What happens to your addresses if you switch providers?
If you’re in marketing, this affects you more than it might seem. A campaign depends on landing pages loading, integrations working, and checkout staying up. Address planning belongs on the launch checklist, next to load testing and monitoring.
Your options for getting more capacity
What makes sense depends on how your infrastructure is set up and how much network management your team can handle.
Use your provider’s addresses
If you’re on managed hosting, an ecommerce platform, or standard cloud services, using the addresses your provider supplies is usually easiest. They handle the admin, and you get on with running your business.
The catch is that you’re tied to their pricing, allocation rules, and migration terms. Make sure you understand those before you build anything that relies on a fixed address.
Buy address space
Organisations with steady, long-term needs sometimes buy IPv4 space through a registry-approved transfer. Expect an upfront cost, plus the work of verifying the block, completing the transfer, and putting the addresses to use.
Owning addresses doesn’t give you hosting or connectivity. You’ll still need a network provider that supports them and people who know how to manage them.
Lease address space
Leasing gives you access to existing addresses for a set period without the upfront purchase. Platforms such as IPXO connect address holders with organisations that need capacity, and offer related management services.
Leasing still takes some groundwork. Check that your hosting or cloud environment accepts addresses sourced from outside, since cloud platforms usually require you to prove authorisation before routing them. Read the renewal terms. Check the reputation of the addresses, because a block previously used for spam can cause deliverability problems. And be clear on who handles abuse reports.
Use fewer addresses before you buy more
Before looking for extra addresses, look at how you’re using the ones you have.
Plenty of internal services never need to face the public internet. Components can talk to each other over private networking, with a shared gateway or load balancer controlling what gets in from outside.
This isn’t free, though. Gateways and load balancers have their own charges, and shared components need enough capacity and redundancy to avoid becoming a single point of failure. On AWS, for instance, NAT gateways are billed by the hour and for every gigabyte they process. Cutting your address count won’t always cut your total bill.
The aim is the right design for the workload. Fewer addresses only help if the system stays reliable, secure, and affordable to run.
Make room for IPv6
IPv6 uses 128-bit addresses, so it has so many that scarcity stops being a concern. The RIPE NCC has been clear that stopgaps like transfers and address sharing don’t fix the underlying shortage, and that IPv6 is the real answer.
Switching isn’t a one-off job. Hosting, applications, monitoring, security tools, and third-party integrations all need to support it. Most public-facing services will need to run both protocols for a while, so customers stuck on IPv4-only connections can still reach you.
An easy place to start is to make IPv6 support a standard question when you evaluate suppliers or scope new development. That way you avoid adding systems you’ll have to rework later.
Leasing, buying, and using addresses more efficiently all solve today’s problem. IPv6 is what solves it for good.
Build address planning into growth
IPv4 scarcity won’t stop a digital business from growing. It does make your infrastructure choices, supplier terms, and resource use matter more.
As a business owner, you should know who manages your addresses, where those costs appear, and whether an upcoming launch will need more. Get marketing and technical teams talking about it before anyone commits to a schedule.
The practical approach has three parts: use what you have efficiently, have a plan for getting more IPv4 when you need it, and move towards IPv6 over time. Do that, and your infrastructure spending will track what the business actually needs.