VegaMSP
← All articles Optimize Network Bandwidth for Teams Voice Traffic How-to

Optimize Network Bandwidth for Teams Voice Traffic

Table of Contents

Last Updated: October 17, 2026

Assess Your Network for Microsoft Teams Voice Readiness

Network readiness is the first thing to check before you push Teams voice traffic across your business. A network assessment catches problems before your team ever picks up a headset. Skipping this step is a common mistake, and it shows up later as dropped calls and frustrated staff.

A Microsoft Teams network assessment checks four things:

  • Available bandwidth on your internet connection and each network route
  • Latency, jitter, and packet loss from your office to the cloud endpoint
  • Firewall rules and port ranges for real-time media
  • Wi-Fi coverage in the rooms where people actually take calls

Run the Microsoft 365 network connectivity test from a few desks, not just the server room. Check results during peak usage, not at 8 a.m. when the office is empty. Document what you find so you can measure improvement later.

Watch Out Testing only from a wired desk in the IT office hides the real problem. Most voice quality complaints come from Wi-Fi dead zones and congested switches, so test where your people sit.

Microsoft Teams Bandwidth Requirements for Voice and Video

Microsoft Teams bandwidth requirements are modest for audio and heavier for video, but the numbers only help if you translate them into a plan for your actual team.

The mistake most teams make is sizing to headcount instead of concurrency. If 40 people are on the phone at once, you size for 40 simultaneous streams, not 250 employees.

Here is a worked planning worksheet you can fill in with your own numbers:

Traffic Type Per-Stream Estimate Your Peak Count Total
Audio call ~100 Kbps each way ___ concurrent calls ___ Kbps
HD video call ~1.5-2.5 Mbps each way (varies) ___ concurrent calls ___ Kbps
Screen sharing Varies by content ___ sessions ___ Kbps
Normal office traffic Your measured baseline n/a ___ Kbps
Headroom 30-50% buffer n/a ___ Kbps

Worked example for a 200-person office. Assume 60 concurrent audio calls (6 Mbps each way), 20 concurrent HD video calls (40 Mbps each way), 10 screen-sharing sessions (10 Mbps each way), and a 60 Mbps office baseline. That totals roughly 116 Mbps each way before headroom.

A few planning rules that hold up in practice:

  • Size to concurrent streams, not seats. Concurrency rarely exceeds 20-30% of headcount for audio, but video concurrency can spike during all-hands and training sessions.
  • Watch the upload direction. Most consumer and small-business circuits are asymmetric, and voice plus video plus screen sharing stresses upload first.
  • Keep utilization under 70% at peak. Above that, queuing delay and jitter climb even with QoS in place, because QoS prioritizes packets but cannot create capacity.
  • Re-measure quarterly. Headcount, video adoption, and cloud app usage all drift upward, and last year's plan quietly becomes this year's bottleneck.
Watch Out Bandwidth planning on paper is not the same as bandwidth delivered in practice. Measure actual throughput during peak hours from the desks where people sit, not from the server room, because Wi-Fi contention and switch congestion can cut effective capacity well below what the circuit advertises.

If your total peak demand eats more than 70% of your available capacity in either direction, you have a problem waiting to happen. Either add capacity, reduce concurrency through scheduling, or move video-heavy meetings to a time when the link is quieter.

Implement Microsoft Teams QoS Configuration

Quality of Service (QoS) marks voice and video packets so your network equipment handles them first. Microsoft Teams QoS configuration starts with tagging media traffic with a DSCP value, then telling your switches and routers to honor those tags.

Here is the practical sequence:

  1. Set the DSCP values for audio and video in your Teams admin policy
  2. Configure your switches to trust and preserve those markings
  3. Set up queues on your routers for each traffic class
  4. Verify the markings survive the trip from desk to internet
An IT professional configuring network settings on a laptop in a server room, with network cables and switches visible in the background
An IT professional configuring network settings on a laptop in a server room, with network cables and switches visible in the background

Most Teams voice problems trace back to QoS markings that get stripped somewhere in the path. A switch that resets DSCP to zero undoes the whole setup. Test each hop.

Pro Tip If your team is spread across multiple offices, document the DSCP values and queue settings for every network vendor you run. Mixed equipment is where prioritization quietly breaks.

Prioritize Teams Voice Traffic with Port Ranges and Firewall Rules

Traffic prioritization depends on clean firewall rules and the right port ranges. Teams media traffic uses UDP for real-time audio and video, with TCP as a fallback. Block those ports and calls fail or fall back to lower quality.

Your firewall rules should allow:

  • UDP and TCP media ports for Teams real-time media
  • Outbound access to Microsoft's cloud endpoints
  • Return traffic without deep inspection that adds delay

Microsoft publishes the current IP ranges and port details in its official documentation. Check them before you write rules, because the ranges change over time. Microsoft Teams network planning guidance

If your team works remotely over a VPN, consider split tunneling. Send Teams media traffic straight to the internet instead of through the VPN tunnel. Hairpinning voice through a distant VPN gateway adds latency and jitter that no amount of bandwidth fixes.

Diagnose Teams Packet Loss, Latency, and Jitter

Teams packet loss, latency, and jitter are the three numbers that explain almost every call quality complaint, but defining them is not the same as diagnosing them. The gap most guides leave open is a repeatable method: how to measure, what thresholds matter, and how to tell which of several possible causes is actually responsible. This section gives you that method.

Get Started Today →

The three metrics and what they mean for voice:

  • Latency is one-way delay between the user's device and the Teams cloud endpoint. Under about 100 ms one-way is generally comfortable; sustained delay above 150 ms makes conversations feel laggy and causes people to talk over each other.
  • Jitter is variation in delay between packets. Under about 30 ms is generally tolerable; higher jitter makes audio chop and cut out because the jitter buffer cannot smooth the stream.
  • Packet loss is missing data. Even 1% loss can be audible on voice; sustained loss above 2-3% breaks call quality noticeably, and bursty loss is worse than evenly spread loss.

A repeatable before-and-after test plan:

  1. Pick test points that match where people sit. Test from at least three desks per floor, including a meeting room and a Wi-Fi-only location, plus one wired desk as a control.
  2. Test during peak usage, not at 8 a.m. Run the Microsoft 365 network connectivity test and a sustained ping or UDP probe to the Teams endpoint during your busiest hour.
  3. Capture a baseline before changes. Record latency, jitter, and loss for at least 10 minutes per test point so you have a distribution, not a single sample.
  4. Make one change at a time. Adjust QoS, Wi-Fi, or routing, then re-run the same tests at the same points and times.
  5. Compare against thresholds, not against feelings. A change that moves jitter from 45 ms to 20 ms is a win even if users have not complained yet.

A diagnostic decision tree for poor voice quality:

  • If one office reports problems and others do not: the issue is local. Check Wi-Fi coverage and roaming in that office, switch uplink congestion, and whether QoS markings survive the local switches.
  • If everyone reports problems at the same hour: the issue is capacity or upstream. Check internet circuit utilization, peak concurrency, and whether QoS is actually shaping traffic on the WAN edge.
  • If calls are choppy but web browsing feels fine: suspect jitter or Wi-Fi interference rather than raw bandwidth. Web traffic tolerates jitter; voice does not.
  • If calls drop or fail to connect: suspect firewall or port blocking, not bandwidth. Verify UDP media ports and outbound access to Microsoft's endpoints.
  • If audio is one-way or delayed by seconds: suspect routing or VPN hairpinning. Trace the path from the device to the Teams endpoint and look for asymmetric routes or a distant VPN gateway.
  • If only remote users are affected: suspect split tunneling and home network conditions. Compare their path to the office path.

Use the Teams call quality dashboard and per-user call analytics to confirm patterns before you change configuration. The dashboard tells you what users experienced; your own probes tell you why.

Key Takeaway Measure from the user's device to the cloud endpoint, not from your firewall. The gap between those two points is where most voice quality problems hide, and a decision tree that separates local, upstream, and endpoint causes will save you from fixing the wrong thing.

Run this validation cycle before any major change and again after, and keep the results. A documented before-and-after is the difference between guessing and knowing whether your optimization actually worked.

Optimize Wi-Fi and Local Network for Teams Voice

Wi-Fi is where voice quality goes to die in most offices. A strong signal bar does not mean a good call. Voice needs steady, low-jitter delivery, and Wi-Fi delivers that only when it is designed for it.

Start with these fixes:

  • Add access points where coverage is weak, especially meeting rooms
  • Use the 5 GHz band for voice devices and keep crowded 2.4 GHz for guests
  • Separate voice traffic onto its own network segment where you can
  • Cap guest and IoT devices so they cannot crowd out calls

Wired connections still beat Wi-Fi for desk phones and docking stations. If a role involves back-to-back calls, give that person a cable. It is the cheapest call reliability upgrade you can make.

Plan for Peak Usage and Build Resilience for Teams voice

Peak usage planning is the difference between a network that survives Monday morning and one that folds. Map your busiest hour, then add buffer on top of it.

Build resilience with redundancy:

  • A backup internet connection so a single outage does not stop calls
  • Failover routing that moves voice traffic automatically
  • Monitoring alerts that flag congestion before users complain

Review your bandwidth capacity every quarter. Headcount grows, video use grows, and last year's plan quietly becomes this year's bottleneck.

Pro Tip Run a peak-load test before a big all-hands or busy season. Push concurrent calls to your expected maximum and watch latency and jitter in real time. You will find the weak link while it is still cheap to fix.

Conclusion: Validate and Maintain Your Teams Voice Network

Networks drift. A configuration that delivered clean calls in January can degrade by summer as devices, apps, and headcount pile on. Validation is not a one-time task.


Getting Teams voice right takes steady measurement, clean QoS configuration, and someone watching the numbers. That is exactly the kind of work VegaMSP handles for growing teams. Our fully managed network services keep bandwidth and prioritization tuned, our VoIP integration keeps calls running on a network built for them, and unlimited helpdesk support means you are never stuck troubleshooting alone. Get started with VegaMSP and run Teams voice on a network that helps eliminate downtime.

Frequently Asked Questions

How much bandwidth does Microsoft Teams voice traffic need?

Microsoft Teams voice calls use about 100 Kbps per call for audio. For video calls, bandwidth ranges from 150 Kbps to 1.5 Mbps depending on resolution. Meetings with screen sharing can use up to 2.5 Mbps. To ensure smooth performance, Microsoft recommends planning for 1.5 Mbps per concurrent user for voice-only and 4 Mbps for video. Always add a 20% buffer for overhead and peak usage.

How do you prioritize Microsoft Teams voice traffic on a network?

Prioritize Teams voice traffic by implementing Quality of Service (QoS). First, identify Teams media traffic using port ranges (UDP 3478-3481) or application recognition. Then, configure network devices to mark voice packets with DSCP code 46 (EF) and place them in a high-priority queue. Ensure that all routers and switches along the path honor these markings. This reduces latency and jitter during congestion.

What QoS settings should you use for Microsoft Teams calls?

For Teams calls, use DSCP markings: Audio should be marked EF (46), video AF41 (34), and sharing AF21 (18). Configure your network to trust these markings from Teams clients and apply priority queuing. Microsoft recommends enabling QoS on all Windows clients via Group Policy and on network devices. Avoid over-prioritizing to prevent starvation of other traffic.

How can you reduce packet loss, latency, and jitter in Teams calls?

Reduce these issues by ensuring sufficient bandwidth, implementing QoS, and minimizing network hops. Use wired connections instead of Wi-Fi when possible. Configure split tunneling to avoid VPN backhaul for Teams traffic. Monitor network performance with tools like the Microsoft 365 network connectivity test. Address any congestion points by upgrading links or adjusting QoS policies. Target latency under 100 ms, jitter under 30 ms, and packet loss below 1%.