Before You Call Your ISP
Something's wrong with your internet. Pages are loading slowly, your video calls keep freezing, or maybe everything just feels off. Before you spend an hour on hold with your ISP, there's a good chance you can diagnose the problem yourself — and potentially fix it in minutes. This network diagnostic guide walks you through the process from simple to advanced.
Step 1: The Basics (2 Minutes)
Start with the obvious stuff. It's surprising how often these resolve the issue:
- Restart your router and modem. Unplug both, wait 30 seconds, plug the modem back in first, wait for it to fully boot, then plug in the router. This clears temporary state and fixes a surprising number of issues.
- Check your cables. Make sure ethernet cables are firmly seated. Look for damaged or bent connectors. If you're on Wi-Fi, move closer to the router and test again.
- Try a different device. If your phone works fine on Wi-Fi but your laptop doesn't, the problem is your laptop — not the network.
Step 2: Run an Internet Diagnostic Test
A speed test is your first real data point. Use a reliable tool and pay attention to all three metrics:
- Download speed: Should be close to what your plan advertises (within 70–80% is typical).
- Upload speed: Especially important if you're having issues with video calls or uploads.
- Latency (ping): Should be under 30 ms for most connections; under 10 ms is excellent.
Run the test multiple times at different times of day. If speeds vary wildly, congestion (yours or your ISP's) may be the issue. Test on both Wi-Fi and a wired ethernet connection to isolate whether the problem is your wireless setup.
Step 3: Ping and Traceroute
These two commands are the bread and butter of network diagnostics.
Ping Test
Open your terminal or command prompt and run:
ping 8.8.8.8
This sends packets to Google's DNS server and reports the round-trip time. Look for:
- Average latency: Should be consistent and low.
- Packet loss: Even 1–2% is a problem. Run
ping -n 100 8.8.8.8for a larger sample. - Consistency: If latency jumps from 15 ms to 200 ms randomly, you have a stability issue.
Traceroute
Run a traceroute to see where delays occur in the path to a destination:
tracert 8.8.8.8 # Windows
traceroute 8.8.8.8 # macOS / Linux
You'll see a list of routers (hops) between you and the destination. If one hop shows a massive latency spike, that router is the bottleneck. A few high-latency hops at the end are normal (some routers deprioritize diagnostic packets), but a spike in the middle of the route is cause for concern.
Step 4: DNS Check
If websites load slowly but your speed test looks fine, DNS might be the culprit. Your ISP's DNS server could be slow or unreliable.
Test DNS resolution speed:
nslookup speedchecktest.com
If it takes more than a few hundred milliseconds, try switching to a public DNS provider:
- Cloudflare:
1.1.1.1and1.0.0.1 - Google:
8.8.8.8and8.8.4.4 - Quad9:
9.9.9.9
Change DNS in your router's settings to apply the fix to every device on your network.
Step 5: Wi-Fi Interference Check
Wi-Fi performance is heavily affected by interference. Common culprits:
- Neighboring networks: Use a Wi-Fi analyzer app to see which channels are congested. Switch to a less crowded channel (1, 6, or 11 for 2.4 GHz; 5 GHz channels have more room).
- Physical obstacles: Walls, floors, metal appliances, and even fish tanks degrade Wi-Fi signals. A mesh system or wired access points can help in large homes.
- Older devices: A Wi-Fi 4 (802.11n) device on a Wi-Fi 6 router will bottleneck your speeds.
Step 6: Advanced Diagnostics
Bufferbloat Test
Visit a bufferbloat testing site (like DSLReports or Waveform's bufferbloat test) while your connection is under load. If your grade is C or below, your router's buffer management is poor. Enabling SQM (Smart Queue Management) or fq_CoDel on your router can dramatically improve performance under load.
MTU Discovery
Incorrect MTU (Maximum Transmission Unit) settings can cause fragmented packets and reduced throughput. Test with:
ping -f -l 1472 8.8.8.8 # Windows
ping -D -s 1472 8.8.8.8 # macOS / Linux
If you get "message too long" errors, reduce the number until it works. Add 28 to get your optimal MTU. Standard is 1500; some ISPs use 1492 (PPPoE) or other values.
Connection Monitoring
For persistent issues, monitor your connection over time. The Windows command pathping 8.8.8.8 combines ping and traceroute, sampling over a longer period. On Linux/macOS, tools like mtr provide real-time hop-by-hop monitoring. This helps catch intermittent issues that a single test would miss.
When to Call Your ISP
You've done the diagnostics. You know your internal network is fine (wired test shows good speeds, no packet loss, low latency). But when you test directly from the modem — bypassing your router entirely — the problems persist. That's when you call your ISP with specific data: your test results, the times when problems occur, and the diagnostics you've run. Armed with that information, you'll get past the tier-1 script readers and straight to someone who can actually help.