Right now, in a typical home, a phone, a television, a game console, and a work laptop may all be online at the same time. Ask a simple question: how many internet addresses does that household actually own? Many people assume the answer matches the number of devices.
It usually does not. Most homes share a single public address, and somehow nothing gets mixed up. Understanding how that works explains a fair amount about privacy, troubleshooting, and why certain services on the internet simply refuse to connect.
One building, one street address, many apartment numbers
Think of an apartment building with a single street address and hundreds of flats inside. Mail arrives at the front door addressed to the building. A person at the reception desk reads the flat number, checks a list, and carries each envelope to the right door.
Nobody outside needs to know the internal layout. They only need the street address, and the reception desk handles the rest.
From the street, the building looks like one location. Inside, it is a small city.
Private numbers and a translation table
That is close to how home networking tends to work. Devices receive private addresses, which only mean something inside the local network. These often look like 192.168.x.x, a range set aside for internal use, so the same numbers can repeat in millions of homes without any conflict.
When a device wants to reach the internet, the router rewrites the outgoing packet. It swaps the private source address for the single public address your provider assigned, and it records that swap in a translation table, together with a port number that marks which internal conversation the reply belongs to.
The technical name for this arrangement is network address translation, usually shortened to NAT. In effect, the router plays the role of that reception desk, remembering which envelope belongs to which flat.
When a reply arrives, the router consults the table, finds the matching entry, and forwards the packet to the correct device. Entries typically expire after a period of inactivity, because holding every record forever would waste memory without any benefit.
You can watch the table at work if you open a router status page, which often lists active translations and the devices behind them. It is a surprisingly readable window into what a household is doing online.
Why the same pattern appears in offices
The pattern shows up far beyond homes. A company with a thousand employees may present only a handful of public addresses to the internet, with internal addressing and translation handled at the network edge. This keeps address usage manageable and gives administrators a natural place to apply policy.
It also means logs and alerts often record the gateway rather than an individual machine, so security teams need extra context from internal systems to reconstruct what happened. Some organizations consolidate that context instead of stitching together separate consoles. Tools like Ping64, an all-in-one office security platform, combine multiple security controls in one system to simplify corporate data protection.
The interesting detail is that translation is a plumbing decision, not a security feature, even though the two are frequently discussed as if they were the same thing. That distinction matters when someone promises protection that a router was never designed to provide.
What translation does not do
A common belief is that NAT hides devices and therefore protects them. That view is more generous than the facts support. Translation makes inbound connections awkward, which does deter some casual scanning, but it is not a filter that inspects traffic, and it does not stop malware from making outbound connections.
Many intrusions begin with a device inside the network reaching outward, and once that connection exists, the reply path is already open. Firewalls, endpoint controls, and prompt patching do the actual protective work.
Another misconception is that NAT and a firewall are the same thing because both live in the same box. They are separate functions that happen to share hardware, and a router may include one, both, or neither in any meaningful configuration. Keeping the two ideas apart may help when you read a specification that treats them as interchangeable.
There are practical limits as well. Some applications, particularly peer-to-peer tools and certain voice services, struggle behind translation and lean on workarounds. Providers also apply translation at a larger scale, which can make hosting a service at home harder and can complicate how abuse is traced later.
IPv6 shifts the picture by offering enough addresses to give devices their own public identity, which reduces the need for translation. Adoption is uneven, so the older arrangement is likely to stay with us for years.
The short version
One public address can serve an entire household because a router keeps a table and rewrites traffic as it passes through. It is bookkeeping, not a shield. Knowing the difference may save you an afternoon of confused troubleshooting, and it makes conversations about privacy and network design a little more honest.
References
- NIST SP 800-41 Rev. 1, Guidelines on Firewalls and Firewall Policy: https://csrc.nist.gov/pubs/sp/800/41/r1/final
- CISA, Secure Our World: https://www.cisa.gov/secure-our-world
- MITRE ATT&CK, T1090 Proxy: https://attack.mitre.org/techniques/T1090/
- ENISA Threat Landscape: https://www.enisa.europa.eu/publications/enisa-threat-landscape-2024



Top comments (0)