A video call freezes just as you start speaking. A game shows a stable download speed but reacts a second late. A remote desktop session feels like every click has to travel across the country and back. These are classic reasons to learn how to troubleshoot high latency – but the fix depends on where the delay is being introduced.
Latency is the time it takes for data to travel from your device to a destination and for a response to return. It is usually measured in milliseconds (ms). Unlike bandwidth, which is the amount of data your connection can carry, latency is about delay. A fast 1 Gbps connection can still feel slow if packets take too long to arrive.
The practical goal is not simply to find a high number. It is to determine whether the delay starts on your device, inside your home or office network, with your ISP, or at the remote service you are trying to reach.
Start by Measuring the Right Kind of Latency
Before changing router settings or calling your ISP, collect a few measurements. Test while the problem is happening if possible, since an idle network can look healthy even when it becomes congested at night or during work hours.
A ping test sends small packets to a target and measures the round-trip time. Test at least two destinations: your router or default gateway, then a reliable public destination. A high ping to your router points to a local problem. A low router ping but high public ping suggests an issue beyond your local network.
For many wired home connections, a ping below 20 ms to a nearby public endpoint is common, though location and ISP routing matter. Wi-Fi commonly adds a little more delay. Consistently high results, sharp spikes, or packet loss are more useful warning signs than a single number.
Also watch for jitter. Jitter is variation in latency from one packet to the next. A connection that alternates between 20 ms and 200 ms can be more disruptive to gaming, voice calls, and remote work than one that stays at a predictable 70 ms. Real-time applications need consistent delivery as much as they need low delay.
Run a speed test as well, but do not treat its download result as a latency diagnosis. It can reveal a separate issue, such as an overloaded connection or poor upload capacity. Pay attention to latency reported when the connection is idle and when it is under load. If latency rises dramatically during a download or upload, your connection may be experiencing bufferbloat – excessive queuing in a router, modem, or ISP device.
How to Troubleshoot High Latency on Your Local Network
Your first job is to separate Wi-Fi and device problems from an internet-path problem. This avoids spending time on ISP troubleshooting when a weak wireless signal is the real cause.
Test with an Ethernet connection
Connect the affected computer or console directly to the router with an Ethernet cable, then repeat the same ping and application test. If latency improves substantially, the internet connection is probably not the main problem. Focus on Wi-Fi coverage, interference, or router placement.
Wireless signals compete with walls, appliances, neighboring networks, Bluetooth devices, and distance. The 2.4 GHz band travels farther but is often crowded. The 5 GHz band usually offers more capacity and less interference over shorter distances. Wi-Fi 6 and 6E can help in busy environments, but they cannot overcome a weak signal through several dense walls.
Move closer to the router, use the 5 GHz or 6 GHz band when available, and keep the router in an open, central location. Avoid placing it inside a cabinet, behind a television, or beside equipment that generates interference. If one far room is consistently affected, a properly placed mesh node or wired access point is often more effective than repeatedly rebooting the router.
Check whether one device is consuming the connection
High latency often appears when another device is uploading cloud backups, syncing large files, streaming high-resolution video, downloading games, or running a security camera feed. Upload activity is especially likely to create lag because many residential plans have much less upload bandwidth than download bandwidth.
Pause large transfers and test again. Check computers, phones, game consoles, smart TVs, and cloud-storage applications. If latency returns only when the network is busy, enable quality of service (QoS) or smart queue management on a router that supports it. These settings prioritize time-sensitive traffic or control queues so one transfer does not delay everything else.
QoS is not magic. It cannot create bandwidth that your service plan does not provide, and incorrect bandwidth settings can reduce performance. Set the available upload and download values slightly below the speeds your connection can reliably sustain, then test under normal use.
Restart carefully and update selectively
Restarting the modem and router can clear a temporary fault, renew a connection, or recover from a software issue. Power off both devices, wait briefly, start the modem first, and then start the router after the modem reconnects.
Do not rely on restarts as a permanent fix. If the issue returns frequently, check router firmware, inspect Ethernet cables, and verify that the router is not overheating. Older routers may struggle with many active devices, fast service tiers, VPN encryption, or advanced traffic-management rules.
Use Traceroute to Find Where Delay Begins
Traceroute shows the network hops between your device and a destination. Each hop is typically a router operated by your ISP, a backbone provider, a content network, or the destination network. It helps narrow down where latency starts increasing.
If your first hop – usually your router – is slow, investigate the local network. If the first few ISP hops show a sustained jump and every destination is affected, the issue may be within your ISP’s access or regional network. If the route stays normal until it reaches a distant network, the destination’s location or its provider’s routing may be the limiting factor.
Interpret traceroute results carefully. Some routers deprioritize or ignore traceroute responses while forwarding normal traffic efficiently. A single hop with a high time or an asterisk does not prove that hop is broken. Look for increased latency that continues through later hops, along with real symptoms in your application.
Try more than one destination. A game server, business application, or website may be geographically distant by design. Data cannot travel instantly across long distances, even on fiber. If one service has high latency while other nearby services are normal, the problem may be server load, a specific route, or the service’s own network rather than your connection.
Rule Out DNS, VPN, and Application Issues
DNS usually affects how long it takes to begin connecting, not the ongoing ping once a connection is established. Slow or failing DNS can make a website appear unresponsive before it loads. If names take a long time to resolve but direct connections work normally, test a different DNS resolver.
VPNs and proxy services can add latency because traffic is encrypted and routed through another server before reaching its destination. Test briefly with the VPN disconnected if your work or security policy allows it. If the delay disappears, choose a closer VPN endpoint or review whether split tunneling is appropriate for your setup.
For gaming, confirm that you selected the correct region. For remote desktops and cloud applications, use the closest available data center. For a server you manage, test from another network and review CPU load, interface errors, firewall rules, and application response times. Network latency and slow server processing can feel identical to the person waiting for a page or command to respond.
When to Contact Your ISP
Contact your ISP when you can show that the issue is outside your local network. Useful evidence includes repeated wired ping tests, timestamps, packet-loss results, traceroute output, and confirmation that the issue affects multiple devices. Your public IP address can also help identify the connection or account details the provider needs, although an IP location result is approximate and does not identify your exact home address.
Describe the pattern rather than saying the internet is slow. For example: latency rises from 25 ms to 250 ms every evening between 7 and 10 PM on a wired connection, with packet loss affecting multiple destinations. That gives support a clearer starting point and can help distinguish neighborhood congestion from a modem signal problem.
If your ISP finds no fault, ask whether the modem is approved for your service tier, whether signal levels are within range, and whether there are known outages or maintenance events. Cable, fixed wireless, DSL, fiber, and satellite connections have different failure patterns, so the best next step depends on the access technology.
High latency becomes manageable once you test one layer at a time. Start close to your device, establish a wired baseline, compare destinations, and save evidence before changing several settings at once. That process turns random lag into a specific problem you can fix, escalate, or work around.
