You Bought a New Router and Nothing Changed. Now What?

The old router is in a drawer, every device is back on the network, and the back bedroom is exactly as bad as it was last week.

That outcome is common, and it rarely means you bought the wrong model. It means the thing that was broken was never the thing a router replaces. Here is how to find what you actually changed, what quietly canceled part of the upgrade, and which purchase would have fixed the room you cared about.

A router upgrade buys capacity, not reach#

Routers are sold on numbers that describe how much traffic they can move at once: aggregate throughput, stream counts, device limits, channel width, band count. None of those describe how far a signal travels.

Reach is decided by three things, and a new box changes none of them. Transmit power is capped by regulators and is effectively identical across consumer routers. The frequency you use sets how well the wave survives a wall. And the walls themselves do not care what year the router was made. Feature by feature, that argument is in whether Wi-Fi 6 or 7 really has better range.

So the upgrade did something. It just did it where you were not looking. A modern router lifts throughput in the rooms that already had a strong signal, and leaves a room at the far end of the house on the same weak signal it had before. If the living room went from 200 Mbps to 600 Mbps while the bedroom stayed at 12 Mbps, nothing regressed. The gap widened, and a wider gap feels like a downgrade.

First, find out what did improve#

Measure before you decide the money was wasted, because the result changes what you buy next. Fifteen minutes with one device is enough.

1. Fix your variables. One device, one speed test, one spot per room. Stand in the same place each time — mark it if you have to. A number you can repeat beats a number that is merely impressive.
2. Test beside the new router. Record download, upload and idle latency. This is the ceiling your house can deliver over Wi-Fi.
3. Test in the room that made you buy it. Same device, same test, within a couple of minutes.
4. Read the actual signal level in both places. Bars round a 60-point scale into four steps. Android shows dBm in the network details, macOS in Wireless Diagnostics, and every router client list reports each device signal — use that for iPhones.
5. Open the client list and look at what each device negotiated. You are looking for anything still on 2.4 GHz, and for devices sitting at very low link rates.
6. Plug a laptop into the router with a cable and test again. If wired is also slow, stop here. The bottleneck is upstream of every radio in the building, and the ten-minute test that decides whether to call your ISP is where to go next.

Fill in three rows and the diagnosis usually writes itself.

A big near-router number, a small far-room number, and 20 dB or more between the two signal readings means the upgrade worked as designed and your problem is coverage. A near-router number that still disappoints means something else is capping the whole house — the next section covers the usual suspects.

Six things that quietly erase the upgrade#

Rule these out before you conclude anything. Each one makes a good router behave like the old one.

1. It went back in the same closet. New routers usually go exactly where the old one lived, which is wherever the coax or fiber enters — often a hallway cabinet at floor level behind a metal panel. That spot costs signal in every direction at once. Pulling the router out of that corner and up onto a shelf typically recovers 5 to 15 dB in every direction, for the price of a longer cable. Work through where to put your router first.

2. The old router is still broadcasting. If your ISP gave you a combined modem-router and you added your own box behind it, you now have two Wi-Fi networks and probably two layers of NAT. Half your devices may still be joined to the old radio in the closet. Look for the old network name in the Wi-Fi list; if it is there, either disable the gateway's radios or put it in bridge mode so the new router does the routing.

3. Your devices never moved to the better band. Cloning the old network name and password means phones rejoin without noticing anything changed — including rejoining 2.4 GHz, which typically tops out around 60 to 90 Mbps of real throughput on ordinary home gear. Forget the network on the devices that matter, rejoin, then confirm in the client list which band each one landed on.

4. Something in the chain is narrower than the router. A 1 Gbps WAN port cannot carry a 1.2 Gbps plan. An older cable modem, an ISP-supplied ONT, or a forgotten 100 Mbps switch left in the path will cap everything downstream of it at roughly 94 Mbps no matter what the router is capable of.

5. It defaulted to a very wide channel. New routers ship configured for maximum headline speed: 160 MHz, sometimes 320 MHz. Each doubling of channel width brings in about 3 dB more noise, so wide channels are excellent in the same room and worse two rooms away. Many of them are radar-shared, so the router must stop and re-check the channel when it hears a radar pattern, dropping devices for anywhere from seconds to minutes. Set 80 MHz on a non-radar channel and see whether the far room and the dropouts both improve.

6. Your devices are two generations behind the router. Every link runs at the slower end's capability, so a household of older phones, tablets and laptops keeps getting older-generation performance from a brand-new box. The headline features need clients that speak them, and a mixed household collects a fraction of the advertised gain.

Where the bottleneck really is#

Match what was wrong before the purchase against what a router can actually change.

If your problem sits in the bottom half of that table, the router was never going to help, and why Wi-Fi is slow in one room only will find the real cause faster than more shopping.

The change that would have worked#

Assume the audit says coverage. Three options, cheapest first, and the third is the one most people buy first.

Move the router. Free, and it improves every room at once instead of one direction. If the coax or fiber entry point traps you, a longer cable — or a MoCA adapter that relocates the router onto existing coax — is cheaper than any new hardware.

Turn the old router into a wired access point. This is the best-value move available to you right now, because you already own two routers. Run a single Ethernet cable, or use the coax already in the wall, put the old unit in access point or bridge mode with the same network name and password on a different channel, and you have a second full-strength radio with no throughput penalty. One wired access point in a sensible spot usually beats three wireless nodes — see wired backhaul with MoCA or Ethernet.

Add a wireless extender or mesh node. The right answer when running a cable is impossible, and it lives or dies on placement. The unit belongs where the router's signal is still healthy — roughly the -60 to -67 dBm window — not in the dead room, which is where almost everyone puts it. The halfway rule and where it breaks has the procedure.

Be honest about the limits. If the problem room reads better than about -60 dBm and is still slow, signal was never the constraint, and an extender only adds another radio to a busy channel. And a home office pushing video calls and large uploads can still miss the bar behind a single-radio extender that repeats every packet on the same channel — that room wants a cable, or a node with a dedicated backhaul radio.

Decide the position before you spend again#

The mistake that produced the disappointing router is repeatable: buying hardware before knowing whether the problem was reach, capacity or the line. A floor plan settles the reach half without buying anything.

That is what Range Up does. Sketch the layout or open one of the starter layouts, set the material preset to what the house is built from, put the router where it actually stands, and read the predicted room-by-room coverage map. Notice what the map does not contain: a field for which router you own. Reach is geometry, and geometry is the half the purchase could not move. Tap the router onto the shelf you were considering, read the map again, and you know whether relocating it beats spending anything at all. If no position rescues the bedroom, mark that room as the one that has to work and let the app weigh spots for a second radio against it, then compare the coverage with and without. It models the house instead of measuring your live network, which is precisely why it can judge hardware still sitting in a store. Map the room that disappointed you, or read how the second-radio decision gets made first.