Quick answer: QoS configuration for VoIP tells your router or switch to treat voice traffic as the highest priority on your network. Without it, file downloads, video streaming, and general browsing compete with your calls for bandwidth — and voice loses every time. The result is choppy audio, jitter, dropped calls, and one-way audio. Proper traffic prioritisation fixes all of these. Specifically, it reserves bandwidth for voice packets so they always move ahead of everything else on the line.
This guide explains exactly how QoS works for VoIP, how to set it up on common hardware, and which mistakes to avoid.
Why VoIP Calls Drop or Sound Bad
Poor VoIP call quality has three main causes. Understanding each one makes the solution much clearer.
Jitter happens when voice packets arrive out of sequence. As a result, audio sounds robotic or broken. Even a small amount of jitter — just 30–50 milliseconds — makes calls noticeably worse.
Latency is the delay between speaking and the other person hearing you. Generally, anything above 150ms becomes noticeable. Anything above 300ms makes natural conversation difficult.
Packet loss means some voice packets never arrive at all. Consequently, words drop out of the conversation entirely. Even 1–2% packet loss degrades audio quality significantly.
In most business networks, these issues don’t come from the internet connection itself. Instead, they come from inside the network — specifically from voice traffic competing equally with everything else. Therefore, QoS configuration solves the problem at the source.
QoS Configuration for VoIP: How It Works
QoS (Quality of Service) is a set of rules your router or switch uses to sort and prioritise network traffic. In practice, it works like a fast lane on a motorway. All traffic shares the same road, but voice packets move to the front of the queue automatically.
Specifically, QoS uses DSCP (Differentiated Services Code Point) markings to tag voice packets as high priority. Your router reads those tags and sends marked packets first, even when the network is busy. Furthermore, it limits how much bandwidth other traffic types can consume when voice is active.
The key point is this: QoS does not add bandwidth. Instead, it manages how existing bandwidth gets distributed. Therefore, it works best when your internet connection already has enough capacity for your call volume — QoS simply ensures that capacity goes to voice first.
Traffic Classes and Priority Levels
Most QoS implementations sort traffic into three or four classes. Each class gets a different priority level and bandwidth allocation.
Class 1 — Voice (highest priority) Voice RTP streams need the lowest possible latency and jitter. Consequently, this class gets absolute priority. No other traffic jumps ahead of it. Typical bandwidth reservation: 30–40% of total uplink capacity for a busy call center.
Class 2 — Signalling SIP signalling packets set up and tear down calls. They need reliable delivery but not the same strict timing as voice audio. Therefore, signalling sits just below voice in priority.
Class 3 — Business applications CRM platforms, cloud dialers, and remote desktop tools go here. In addition, video conferencing traffic often fits this class. These applications need reasonable bandwidth but tolerate slightly more delay than voice.
Class 4 — General traffic (lowest priority) File downloads, software updates, social media, and general browsing sit here. By contrast to voice traffic, these applications handle delays without any impact on the user experience.
Step-by-Step Setup on Common Hardware
The specific menu names vary by router brand. However, the logic stays the same across all of them.
MikroTik Routers
MikroTik handles QoS through its Queues and Mangle rules. First, create mangle rules to mark VoIP traffic — typically UDP traffic on ports 5060 (SIP) and 10000–20000 (RTP). Next, create a Queue Tree with voice packets assigned to the highest priority queue. Finally, set the parent queue limit to match your available uplink bandwidth. MikroTik’s queue system is powerful but requires careful configuration — misconfigured rules can make things worse, not better.
Fortinet FortiGate
FortiGate uses Traffic Shaping Policies to manage QoS. Specifically, you create a traffic shaper for voice traffic, assign it the guaranteed bandwidth and priority level, then attach it to a firewall policy. FortiGate also supports DSCP marking natively, which helps when traffic leaves your network and passes through your ISP’s equipment.
pfSense
pfSense uses ALTQ or HFSC traffic shaping. In pfSense, you define a queue hierarchy under Traffic Shaper, create a queue for VoIP traffic with the highest priority, and then create a floating firewall rule to match SIP and RTP traffic into that queue. The pfSense documentation covers this in detail — follow the official guide rather than relying on older forum posts, since the interface changes between versions.
TP-Link and Consumer-Grade Routers
Most consumer routers include a basic QoS menu under Bandwidth Control or QoS Settings. Generally, these only allow you to prioritise traffic by device IP or application type. For small offices with light VoIP use, this basic prioritisation often makes a noticeable difference. However, for call centers or heavier VoIP deployments, business-grade hardware gives far more precise control.
QoS Configuration for VoIP: Common Mistakes
Even with QoS enabled, call quality problems persist when these mistakes appear in the configuration.
Forgetting to set the correct uplink speed QoS needs to know your actual internet upload bandwidth. If you enter a number higher than your real uplink speed, the router can’t enforce the limits correctly. Consequently, voice traffic still competes with everything else during congestion.
Only applying QoS on the router — not the switches In larger offices, traffic between devices and the switch also needs prioritisation. Therefore, apply DSCP markings at the switch level too, not just at the edge router.
Not marking traffic at the source Ideally, your IP phones and VoIP gateway mark their own packets with the correct DSCP value before the router sees them. If they don’t, your router must identify and mark VoIP traffic itself — which adds processing overhead and can miss some flows.
Leaving WMM disabled on WiFi access points WMM (Wi-Fi Multimedia) is the WiFi equivalent of QoS. Without it, voice packets lose their priority the moment they hit a wireless access point. Specifically, enable WMM on every access point that carries VoIP devices — and verify it actually activates, since some routers show WMM as enabled but don’t apply it correctly.
Applying QoS after the problem gets serious In practice, QoS works best as a preventive measure, not a last resort. Therefore, configure it before call quality degrades — not after your team starts complaining about every second call dropping.
Testing and Monitoring After Setup
After applying the configuration, test before declaring it done.
First, make several calls during peak network usage — when staff are downloading files, streaming, and running cloud applications simultaneously. If call quality holds up under load, the configuration is working.
Next, use a tool like ping or a dedicated VoIP testing tool to measure latency and jitter to your SIP server before and after QoS. Specifically, look for latency under 150ms and jitter under 30ms consistently.
Finally, monitor ongoing performance with a network monitoring tool. Our guide on PRTG vs Zabbix covers two tools that track latency, jitter, and bandwidth consumption across your network continuously.
If QoS is correctly set but call quality still suffers, the problem likely sits upstream — either with your ISP connection or your VoIP provider’s infrastructure. In that case, contact our team and we’ll diagnose whether the issue is inside your network or outside it.
For related infrastructure guides, see our posts on site-to-site VPN (which shares bandwidth with VoIP on many networks) and AMC vs on-demand IT support (for ongoing network maintenance options).
Frequently Asked Questions
What is QoS configuration for VoIP? QoS configuration for VoIP is a set of router and switch rules that prioritise voice traffic over other data on your network. It ensures voice packets move first, which reduces jitter, latency, and packet loss that cause poor call quality.
Does QoS improve VoIP call quality? Yes, significantly — on networks where voice traffic competes with other data. QoS doesn’t add bandwidth, but it ensures existing bandwidth goes to voice first, which eliminates most call quality problems caused by internal network congestion.
Which routers support QoS for VoIP? Most business-grade routers support QoS, including MikroTik, Fortinet FortiGate, pfSense, and Cisco. Consumer routers like TP-Link also include basic QoS. However, call centers and heavier VoIP deployments need business-grade hardware for precise traffic control.
What ports does VoIP use for QoS? SIP signalling typically uses UDP port 5060. RTP voice streams use UDP ports in the range 10000–20000, though the exact range varies by VoIP platform. Your QoS rules should cover both to capture all voice traffic correctly.
What is WMM and why does it matter for VoIP? WMM (Wi-Fi Multimedia) is the wireless equivalent of QoS. It prioritises voice packets over regular data on WiFi networks. Without WMM enabled on your access points, voice traffic loses its priority as soon as it reaches a wireless device — which undoes everything QoS achieves on the wired side.
Started Today
Want to set up a professional call center?
Contact E Tech Solvers for complete solutions.