How to Read Starlink App Statistics: Latency, Packet Loss and Outages
Speed & Performance
A speed test gives you one number at one moment. The Starlink app's statistics are more useful when your connection is freezing, dropping calls, or feeling inconsistent because they help you work out where the problem is occurring.
This guide explains how to read the main signals, build a useful baseline, and choose the next troubleshooting step without chasing every tiny spike. Menu names can vary slightly as the Starlink app changes, but the diagnostic method stays the same.
Start by separating the two halves of your connection
A device reaches the internet across two different links:
- Device to router: Your phone, laptop, TV, or console connects over Wi-Fi or Ethernet.
- Router to the internet: The Starlink terminal communicates through the satellite network.
This distinction matters. A weak bedroom Wi-Fi signal can make Starlink seem slow even when the satellite link is healthy. Equally, perfect Wi-Fi cannot prevent a drop caused by an obstruction or a wider service interruption.
Before diagnosing anything, connect your phone to the correct Starlink network and open the app. If you use a third-party router or bypass mode, some local-network readings or controls may differ.
The readings that matter
Latency
Latency is the time it takes data to make a round trip. Lower and steadier is generally better, but consistency matters as much as a single value.
Brief changes are normal on a live network. Look for repeated jumps that line up with a frozen video call, delayed game response, or slow page load. A lone spike with no noticeable symptom is rarely useful evidence by itself.
Packet loss or ping drop
Packet loss means some data did not complete its journey. Real-time activities such as calls, gaming, remote desktops, and live streaming can react badly to repeated loss even when a download speed test looks good.
Focus on clusters and patterns. If loss appears at the same time as a user-visible problem, it deserves investigation. Occasional isolated events can occur during normal operation and should be judged in context.
Obstructions
The obstruction view records parts of the sky where the terminal's view may be blocked. Trees, rooflines, chimneys, poles, and seasonal growth can all matter.
A repeating interruption pattern is an important clue. If drops happen regularly and the obstruction map shows blocked areas, improve the terminal's view before buying a faster router. Check again after leaves grow, after a storm moves branches, or after any mounting change.
Outage events
The app may group recent events by cause or duration. The wording can change, so pay attention to the pattern rather than one label.
- Obstructed: Investigate the sky view and installation position.
- No signal or network issue: Check whether the event was brief, repeated, weather-related, or shared by other users.
- Searching, booting, or offline: Check power, connectors, and whether the system was intentionally restarted.
- Unknown or uncategorised: Compare the time with symptoms and look for repetition before acting.
Short events that cause no practical disruption are less important than repeated events that match the problem you are trying to solve.
Build a baseline before changing anything
Use the connection normally for at least a day and note how it behaves during quiet and busy periods. A longer observation window is better for intermittent faults.
Record:
- the approximate time of each noticeable interruption;
- which device and application were affected;
- whether the device used Wi-Fi or Ethernet;
- weather conditions;
- any matching latency, loss, obstruction, or outage event;
- whether every device was affected or only one.
This simple log turns “the internet was bad” into evidence you can compare.
A five-step diagnostic test
1. Reproduce the problem
Use the activity that exposes the issue: a video call, online game, large download, or stream. Note the exact time when it fails.
2. Check more than one device
If only one device struggles, update or restart that device and check its Wi-Fi signal. If several devices fail together, the router, terminal, power supply, or wider service becomes more likely.
3. Compare Wi-Fi and wired performance
Where your hardware permits, test a computer over Ethernet. If Ethernet is stable while Wi-Fi is poor, focus on router placement, interference, mesh coverage, or the device itself.
Do not move the Starlink terminal simply because one room has weak Wi-Fi.
4. Match the symptom to the statistics
Open the relevant statistics and outage history. Look at the period around the recorded time.
- Stable satellite statistics, one weak device: Check that device or its local Wi-Fi connection.
- Stable wired connection, poor Wi-Fi on several devices: Check router placement, interference, or mesh coverage.
- Repeated drops matching obstruction data: Inspect the terminal's sky view and mount position.
- All devices disconnect and the app shows the system offline: Check power, cabling, the terminal, router, or service status.
- Slow results mainly at the same busy times: Compare network demand with traffic inside your home.
- Problems only during heavy rain or snow: Investigate weather-related attenuation and installation conditions.
This list is a starting point, not a verdict. Use repeated observations to confirm the pattern.
5. Change one thing and retest
Make one controlled change: move closer to the router, pause a large upload, reseat an accessible cable after safely removing power, or improve an obstruction. Then test the same activity again.
Changing several things at once may fix the symptom, but it prevents you from learning what caused it.
Common mistakes to avoid
Treating one speed test as a diagnosis
Download speed varies with server choice, device capability, Wi-Fi conditions, household traffic, and network demand. Run multiple tests at comparable times and pair them with app statistics.
Confusing Wi-Fi signal with satellite signal
The terminal can have a clear sky while a distant room has poor Wi-Fi. Test beside the router and, if possible, over Ethernet before blaming the dish position.
Reacting to every short event
Statistics contain normal operational noise. Prioritise events that repeat, last long enough to affect use, or align with a recorded symptom.
Factory-resetting too early
A reset can remove network settings and create extra work without fixing obstructions, damaged cabling, power instability, or a wider outage. Capture evidence and try the least disruptive checks first.
Ignoring uploads
Cloud backups, security cameras, large file transfers, and video calls all use upload capacity. Heavy uploads can make latency feel worse for everyone else. Pause them briefly as a diagnostic test.
When to reboot and when to wait
A normal restart can be reasonable when the system is unresponsive or remains offline after you have checked power and accessible connections. Use the app's restart control where available, or follow the instructions for your specific hardware.
After restarting, allow the system time to boot and reconnect. Avoid repeated power cycling. If the statistics show a brief network event that has already cleared, monitoring may be more useful than rebooting.
What to collect before contacting support
Good evidence shortens the troubleshooting conversation. Prepare:
- the time and timezone of recent failures;
- screenshots of relevant statistics, outages, and the obstruction map;
- whether all devices were affected;
- Wi-Fi and Ethernet comparison results, if available;
- recent installation, cable, router, or power changes;
- the troubleshooting steps already tried.
Never post account details, precise home location, or device identifiers publicly.
The practical rule
Use the app to connect a symptom to a pattern. First decide whether the fault is on one device, across the local network, or on the Starlink side. Then make one change and test again.
That approach is slower than guessing for five minutes, but much faster than replacing the wrong equipment.