What you’re paying for when you buy a gigabit plan is capacity, not a dedicated line. It’s a shared pipe into the neighborhood, split across every household pulling from it at the same time.

Think of it as a highway. At 3 AM, the lane is empty and you’re moving at full speed. At 6 PM on a Friday, everyone’s on the road at once, and it doesn’t matter how fast your car is — the traffic jam is the bottleneck, not the engine.

That’s the first thing worth separating out: most people blame “speed” when the actual complaint is latency or packet loss, and those come from a different part of the stack.

Data moves in packets. If a packet is lost, it gets re-requested. A weak Wi-Fi signal drops packets constantly — and the size of your pipe doesn’t matter if the packets never make the trip.

A firehose with a 10-second delay still stutters on a video call, even with huge bandwidth. Downloads care about bandwidth. Real-time calls and gaming care about latency — and a router with plenty of “speed” on paper can still fail both.

Frequency plays into this directly. Higher-frequency bands (5 GHz, 6 GHz) carry more data but travel shorter distances and don’t punch through drywall, furniture, or a floor plan with more than one room between the router and the device. A signal reading 900 Mbps ten feet from the router can drop to a fraction of that one wall away.

Most ISP-provided gateways handle light usage fine — email, browsing, one stream. Once a household runs a handful of devices at once (a TV, a laptop, a console, a stack of smart home sensors), the stock router is doing more work than it was built for.

Mesh systems solve the coverage problem — routing around walls instead of trying to punch through them — but they’re a real cost, typically landing in the $300–$400 range for a multi-node setup. It’s not the first fix to try; it’s the fix for when placement and configuration have already been ruled out.

The router isn’t always the culprit. Oversubscribed neighborhoods — especially newer developments where infrastructure hasn’t caught up to demand — are a real and separate problem.

The isolating test: connect a machine directly to the modem with Ethernet, skip Wi-Fi entirely, and run a speed test. If the wired connection hits the advertised number, the problem lives in the Wi-Fi network. If it doesn’t, the problem is upstream — and that’s a conversation with the ISP, not a router swap.

Wi-Fi trades performance for convenience by design. A wired connection will always beat it on raw consistency — the practical question is how much of that gap actually matters for what’s running in the house.

Further reading: FCC Broadband Speed Guide and the Bufferbloat Project documentation.