Wi-Fi is radio. Ethernet is a cable. The cable does not compete with microwaves, neighbours’ access points, or two walls of brick. That is why a Wi-Fi speed test is often slower than the same plan on a wired PC — and why that gap is useful, not a bug. The tester is not “broken on wireless.” It is showing you the hop that actually sits between you and the internet.

People treat the gap as an insult. The bill says 300 Mbps, the sofa says 40, and someone is accused of theft. In a lot of homes the street is fine. The air is not. Learning to read the gap saves money: you stop buying megabits the radio cannot use, and you start moving the box, changing band, or running a cheap cable to the desk that works all day.

What the gap means

If Ethernet, or a test against the router, is close to your plan and the sofa is not, the ISP line is doing its job. Fix the air. Move the router out of the cupboard. Raise it. Put it nearer the rooms you occupy. Prefer 5 GHz or 6 GHz when the device is close enough to hold the signal. Leave 2.4 GHz for range and for older gadgets that cannot join the faster bands. That is the highest-yield part of improving internet speed at home.

Modern Wi-Fi in the same room can look excellent. The myth that “wireless is always terrible” comes from measuring through walls and then comparing with a one-metre patch cord. Be fair. Compare same tester, two places, same hour. If the same-room wireless is already near the plan, you do not have a crisis. You have physics in the far bedroom.

Why the radio loses

Every wall absorbs energy. Metal, mirrors, fish tanks, and reinforced concrete are rude. The 2.4 GHz band travels farther and is a junk drawer: Bluetooth, microwaves, baby monitors, and every cheap gadget that still ships with a 2.4 radio. 5 GHz and 6 GHz are cleaner and faster, and they fade sooner. An access point that “covers the house” on 2.4 GHz is often covering it with a slow, noisy channel.

Interference is not always a neighbour. Your own mesh, extenders that repeat on the same channel, and a router jammed against a television all count. Extenders that rebroadcast poorly can make the far room look connected and still test like a wet string. A proper extra node on Ethernet backhaul, or a single cable to the desk, usually beats a chain of cheap repeaters.

When both are slow

If the wire is also well below the plan, you are past Wi-Fi. Check other devices, a VPN, then the ISP. Debug the connection in order so you do not replace a router when the street node is congested. Evening-only sag on Ethernet is a different story from a bedroom that is always weak. One is a last-mile conversation. The other is furniture and radio.

Device quality still matters on a cable. A failing USB adapter, an old 100 Mbps port, or a damaged patch cord can mimic a slow ISP. Try another cable before you escalate. Then look at why the tester can sit below the advertised plan if the wire remains short after a reboot.

Honest limits of a browser test

SpeedMetric measures the path from this device to the test server. It does not claim a separate “Wi-Fi speed” unless Wi-Fi is the active connection. Old phones, USB dongles and busy 2.4 GHz networks will look slower than a desktop on Ethernet even when the plan is fine. Compare apples to apples: same tester, two places. That is also how you test internet speed accurately in the first place.

If you only have a phone, the near-router versus far-room pair is still valid. You will not get a laboratory Ethernet number, but you will get the shape of the problem. A USB Ethernet adapter is worth owning if you argue with a provider more than once a year. It turns “my Wi-Fi is bad” into a sentence they cannot dismiss.

Bands, names and sticky phones

Many routers still ship one name for 2.4 GHz and 5 GHz together. Phones then cling to the slower band because it looks “stronger” through a wall. Splitting the names for a week is a cheap experiment: join the 5 GHz name in the same room and test again. If the number jumps, the radio was the story. If it does not, you have a different hop to chase. Steering features on newer systems can do this automatically; measuring still beats trusting the brochure.

Guest networks, IoT VLANs and a microwave on the counter during the test all belong in the notes. A “slow Wi-Fi” evening that coincides with dinner in a small kitchen is sometimes a 2.4 GHz appliance, not a failing provider. Move the test two metres, wait for the oven to finish, and run the tester once more before you buy anything.

Common questions

Is Ethernet always faster than Wi-Fi?

For a given plan, a healthy Ethernet run usually comes closer to the line rate. Modern Wi-Fi can still be excellent in the same room. Distance, walls and interference are what usually open the gap.

My laptop has no Ethernet port. How do I compare?

Test standing next to the router, then in the room where you work. A large drop with no other change is the wireless path. A USB Ethernet adapter is the cleanest second measurement if you have one.

Will switching to 5 GHz close the gap?

Often, when you are close enough. 5 GHz and 6 GHz carry more data and usually see less neighbour noise, but they fade faster through walls. 2.4 GHz reaches farther and is more crowded. Test both instead of guessing.