Broadband speed for Zoom, Teams, and video calls
Calls stay smooth when delay stays predictable. Zoom and Microsoft Teams adapt bitrate up and down, but they hate jitter—latency that keeps changing. A “fast” UK fibre line with choppy timing still produces robotic audio and frozen tiles.
What matters most
| Most important metric | Stable latency and low jitter; audio often fails before video does. |
|---|---|
| Camera on vs off | Camera off cuts upstream demand—a practical fix on busy uplinks. |
| Watch out for | Hidden uploads: cloud backup, photo sync, or a teenager’s stream while you present. |
Why a fast line can still fail calls
Download Mbps measures how wide the pipe is, not how evenly packets arrive. A Wirral home office hits 120 Mbps on Pulse yet Teams still warbles because the laptop is on dodgy 2.4 GHz Wi-Fi and jitter spikes whenever someone microwaves dinner.
Two UK scenarios
Two parents in Cambridge share 9 a.m. stand-ups while a teenager uploads coursework in the next room. The download test looks fine; the call breaks because upstream is saturated—something Pulse alone will not quantify.
A charity worker in Wales runs audio-only on Teams through a storm because video would saturate their uplink. That is not shameful engineering; it is realistic bandwidth triage.
What to try first
- Wired connection for the device that hosts important calls.
- Turn video off or lower quality when audio matters most.
- Pause cloud sync during the call window.
How Pulse relates to this topic
Pulse is useful for checking download capacity and how steady latency and jitter look in your browser session. It does not measure upload, which is half the call story—read does Pulse measure upload? and add another test if upstream is suspect.
Run the Pulse speed test · Read methodology · Review privacy
FAQ
Why does upload matter for calls?
Your camera and microphone travel upstream. If upload is crowded, video freezes even when download Mbps looks generous.
Can I trust a download-only test for meetings?
Pulse helps with download headroom and delay behaviour but does not measure upload—see the upload explainer for limits.
Does turning the camera off really help?
Often yes on marginal links: less upstream and less CPU work on the laptop.
Why do calls stutter when speed tests look fine?
Jitter and bufferbloat style issues do not always show as low Mbps. Stability beats raw speed.