NETCHECK GUIDES · UPDATED 11 SEPTEMBER 2026
How to read your internet speed test
One result is a snapshot. A useful comparison starts with knowing what was measured and keeping the test conditions consistent.
What the numbers mean
| Result | What it tells you |
|---|---|
| Download · Mbps | How quickly test data arrives at your browser. |
| Upload · Mbps | How quickly your browser sends test data that the server acknowledges. |
| Ping · ms | How long a small HTTP request and response take in this test. |
| Jitter · ms | The mean difference between consecutive latency samples in NetCheck. |
Download matters when receiving files; upload matters when sending them. Latency describes delay, not the volume transferred. A larger download number does not automatically mean a more responsive connection.
Further reading: Cloudflare’s explanation of latency, bandwidth and throughput.
Mbps is different from MB/s
Internet packages and NetCheck use megabits per second (Mbps). A file-download app may display megabytes per second (MB/s). There are eight bits in a byte, so a measured 40 Mbps corresponds to a theoretical 5 MB/s before overhead. This conversion is not a promise that every file server will deliver at that rate.
Make a fair comparison
- Keep the same device, browser and test website for the first comparison.
- Pause your own large downloads, cloud backups and uploads. Note whether other people are using the connection.
- Keep the device in the same location. Record whether you used Wi-Fi, Ethernet or mobile data.
- Run the test to completion. Repeat a small number of times, leaving a pause between tests, and compare the typical result rather than only the highest.
- Change one condition, such as moving closer to the router, and repeat. This makes the difference easier to interpret.
How NetCheck measures speed
NetCheck measures latency first, then download and upload. Each transfer direction uses four concurrent HTTP connections over a 12-second window after warm-up. Download counts received bytes; upload counts server-acknowledged bytes. The final value uses transferred bytes and elapsed time, so it can differ from the changing meter during the test.
A 1 GB limit per direction can shorten a very fast test. The meter animation is a visual aid, not an extra measurement. These tests transfer real data: consider your data allowance before repeated testing, especially on mobile networks.
Why two test websites can disagree
Different endpoints, transfer sizes and timing methods measure different paths and conditions. In NetCheck, browser overhead and the short test window are practical limits. If one result is unexpectedly low, compare repeated measurements and your actual app experience before concluding that the provider is at fault.
What to keep for support
Save the time, connection type, results and what was slow. Signed-in NetCheck users can keep recent measurements in the dashboard. If Wi-Fi is the suspected cause, continue with our slow Wi-Fi guide.