How this speed test actually measures your connection
A speed test is only meaningful if it fills your connection the way real use does. A single download stream rarely does, because one connection is limited by latency and packet loss long before it saturates a fast line. This test opens six parallel connections for download and four for upload, which is how a browser loading a heavy page or a download manager behaves, and adds up the bytes that genuinely cross the wire.
Three details do most of the work. The data is incompressible and served with compression disabled, so no gzip layer can inflate the number. The first second or so of every transfer is thrown away, because TCP starts slow and ramps up, and including that ramp would understate a fast line. Finally the numbers you see are payload bytes counted by the browser as they arrive, not an estimate from a file size and a stopwatch.
Mbps and MBps are not the same thing
This trips up almost everyone. Internet plans are sold in megabits per second, written Mbps. File downloads are shown in megabytes per second, written MBps. There are eight bits in a byte, so you divide by eight to go from one to the other. A 100 Mbps plan tops out around 12.5 MB per second, and a 1 Gbps plan around 125 MB per second. When a download sits at 6 MB per second on a 50 Mbps line, nothing is wrong: that is the line running flat out.
What speed do you actually need?
| Activity | Download needed | Also matters |
|---|---|---|
| Web browsing and email | 3 to 5 Mbps | Ping under 100 ms |
| HD video streaming (1080p) | 5 to 8 Mbps per stream | Stable speed |
| 4K video streaming | 25 Mbps per stream | Consistency over peak |
| Video calls (Zoom, Teams) | 3 to 5 Mbps | Upload and low jitter |
| Online gaming | 5 to 15 Mbps | Ping under 40 ms, jitter under 10 ms |
| Working from home | 25 Mbps | Upload of 10 Mbps or more |
| Large file uploads and backups | Any | Upload speed is the bottleneck |
| Busy family household | 100 Mbps and up | Router quality |
Notice how little raw speed most activities need. Beyond about 100 Mbps, extra download speed rarely changes how the internet feels day to day. What you feel instead is latency, jitter and how well the router copes when several devices are busy at once.
Ping and jitter, explained without jargon
Ping is the round trip time for a small packet: out to the server and back, measured in milliseconds. It is the delay before anything starts happening, so it decides whether a connection feels instant or sluggish. Under 30 ms is excellent, under 60 ms is fine for almost everything, and above 150 ms you start to feel the lag in calls and games.
Jitter is how much the ping varies from packet to packet. A line with 25 ms ping and 2 ms jitter is smooth. The same 25 ms ping with 40 ms jitter will stutter in a video call, because packets arrive out of rhythm and the software has to wait or drop them. Jitter is usually the real culprit when a call breaks up on a connection whose download speed looks perfectly healthy. Wi-Fi interference and an overloaded router are the two most common causes.
Why your result may be lower than the speed you pay for
Wi-Fi is nearly always the answer. A connection that delivers 500 Mbps to the router can easily arrive at a laptop two rooms away as 80 Mbps, and older devices with 802.11n radios cap out well below modern plan speeds regardless of signal strength. Before blaming the provider, plug into the router with an ethernet cable and run this test again. The gap between the two runs is the cost of your Wi-Fi.
Other common causes, roughly in order of how often they matter: other people in the house streaming or backing up, a VPN adding an encrypted detour, an old router that cannot route at full line rate, background updates on your own device, and peak evening congestion on the provider's network. Distance to the test server matters too, which is worth remembering because our server is in the UK: a result measured from another continent reflects the route as much as your line.
How to get a result you can trust
Close other tabs and pause any downloads, streaming or cloud sync first, since they compete for the same line. Test on ethernet if you want to know what the line delivers, or on Wi-Fi in your usual spot if you want to know what you actually get. Run it two or three times and take the best result rather than the first, because the first run can be affected by whatever the connection was doing a moment earlier. If you are testing to open a complaint with your provider, run it at different times of day and keep a record: the history table above stays on your device and gives you exactly that.
One honest limitation worth knowing. Every browser based speed test is bounded by the capacity of the server it talks to and the route in between, ours included. If your connection is very fast, treat the result as a floor rather than a ceiling, and compare several tests rather than trusting a single number from any one of them.
Upload speed deserves more attention than it gets
Most plans are asymmetric, so a headline 100 Mbps package might upload at only 10 Mbps. That is invisible while you browse and streams fine, then becomes very obvious the moment you join a video call, back up photos, upload video, or work from a cloud drive. If your work involves sending files rather than receiving them, upload is the number to shop on, and it is the one number that cheap fast-sounding plans usually skimp on.