Google lists up to 3.6 Mbps upload and download per participant for 1080p Meet video, and up to 1.7 Mbps each way for 720p. These are use-case planning figures, not universal minimums or a guaranteed resolution. Also consider concurrent devices, other traffic and the quality options available to your account and hardware.
Google’s published per-participant bandwidth guidance
The following figures are from Google’s network-planning guidance for individuals or small organisations. They describe the stated media use cases, not a promise that an internet plan at the exact figure will sustain every meeting.
| Use case | Outbound / upload | Inbound / download |
|---|---|---|
| 1080p video | Up to 3.6 Mbps | Up to 3.6 Mbps |
| 720p video | Up to 1.7 Mbps | Up to 1.7 Mbps |
| Group meeting | 250 Kbps and above, depending on sent resolution | Up to 4.0 Mbps |
| Audio only | 100 Kbps | 100 Kbps |
Google separately publishes organisation-wide averages and live-stream planning figures. Do not mix those with the per-participant table or interpret “up to” as a universal minimum. Allow additional capacity for other concurrent traffic. Account, hardware and camera capabilities also affect which quality options are available.
Check readiness from the actual meeting location
Use the video-call connection test to check the browser connection before a meeting. It does not join Google Meet or measure its media-server route. For several people at home, use the household bandwidth planner rather than multiplying one requirement by every account invited to a meeting.
- Check both upload and download and pause optional transfers for a quiet comparison.
- Compare Ethernet if available, or the same device nearer the router. A room that performs differently can be recorded with the Wi-Fi room comparison.
- Use Meet’s current video/audio settings to compare a lower workload when quality is poor. A camera preview alone cannot demonstrate the connection to other participants.
- If the issue occurs mainly while files transfer, compare idle and loaded browser timing, then check the effect inside Meet.
Separate network symptoms from a busy computer
A hot or heavily loaded computer can struggle with effects, screen sharing, or multiple apps even while the connection is adequate. Compare a simpler layout or visual workload and observe whether the symptom changes. Do not change graphics, the Wi-Fi band, security settings and the browser all at once; keep the comparison interpretable.
For schools and managed business networks
Ask the administrator to use Google’s network guidance if Meet repeatedly fails only on the managed network. A proxy, firewall or traffic-handling policy needs an organisation-specific review. Do not disable security protection or open broad inbound ports based on a generic household article. Record the time, device, connection, affected feature and the app’s own error or diagnostic information.
Plan two simultaneous Meet calls
As a simple example, two devices using the individual 720p allowance give 2 × 1.7 = 3.4 Mbps in each direction before other traffic and headroom. Two devices using the 1080p allowance give 7.2 Mbps each way. These are planning examples from the table, not measured usage. Invited participants in other households do not each consume another device’s full allowance on your own connection.
Schools should count peak simultaneous endpoints and use the separate organisation-level guidance with their administrator. Do not substitute an organisation-wide average for one device’s quality target.
Check Troubleshooting & help inside Meet
Google’s quality troubleshooting guide describes tailored suggestions under More options → Troubleshooting & help when a quality issue is detected. Record the app’s message and the affected feature, then change one condition and compare.
| Symptom | Comparison to try |
|---|---|
| Others cannot see your outgoing video clearly | Check upload and optional upstream transfers, plus the app’s camera or device suggestions. |
| Problems mainly during screen sharing or effects | Compare a simpler workload and watch device resource use. |
| Problems mainly in one room | Compare the same device nearer the router or on supported Ethernet. |
| It fails only on the managed school or work network | Give the administrator the time, device, error and affected feature. |
Run separate browser speed tests before or after the meeting; a full throughput test adds competing traffic during a call. The slow-internet troubleshooter can help choose the next connection check without assuming the cause.
Frequently asked questions
Is Meet always more bandwidth-efficient than Teams or Zoom?
No universal ranking is established here. Compare the same quality, layout and feature workload, and consult each service’s documentation. Different endpoints and adaptive behaviour can change results.
Will a browser connection test prove a meeting will be smooth?
No. It provides useful baseline evidence. Actual media quality also depends on the service route, both sides’ conditions, the device and the meeting workload.