About & method

How the measurements work

• A connection is not treated as good simply because the phone stays connected or shows LTE/5G. What matters is whether the service people pay for can actually be used, so connection reliability, data performance and responsiveness are reported separately.
• Cell Service - Verification System measures app-layer usability: whether small web requests actually complete, how long they take, and roughly how fast data moves. Connection checks run about every 2 seconds, a download check about every 30 seconds and an upload check about every 60 seconds on a road trip; a stationary spot test measures far more densely.
• Speeds are measured in kbps and stored that way as raw evidence; they are shown in Mbps (kbps / 1000) for readability. Only genuinely independent speed checks are counted - values carried forward onto later measurements are never counted again.
• Data-performance bands (Very Poor / Poor / Fair / Good / Excellent) and the data-impaired threshold are Cell Service - Verification System beta usability thresholds, not regulatory standards, and will be recalibrated.
• A data-impaired observation means the phone WAS connected but an independent speed check measured under 2 Mbps download or under 0.5 Mbps upload. This is deliberately separate from a dead zone, which means nothing got through at all.
• Web and iOS browsers do not expose cellular radio measurements, so those reports contain no dBm/RSRP/RSRQ/SINR signal-strength readings. The optional native Android build may include additional raw radio diagnostics, but these are clearly marked and never treated as proof of usable service on their own.
• A suspected dead zone is 10 seconds or more of continuous failed checks with no app interruption in that window. It stays 'suspected' until several independent trips on separate days see the same area - only then is it centrally described as confirmed.
• Periods where the app was backgrounded or the screen locked are logged as interruptions and excluded from scores, so a paused test is never reported as a dead zone.
• Location comes from the browser's own geolocation, so accuracy varies. Indoor spot tests describe a building or place, never a specific room position.
• Phone call, video/live-stream and browsing results are ESTIMATES from data checks, not measured calls or streams. No actual phone calls or 911 calls are made. Checks run about every 2 seconds, so an outage's exact length (for example 1.5 seconds) cannot be measured precisely - one missed check is treated as a gap. Browsing windows are 30 seconds; call and stream windows are 1 minute of data checks, not real one-minute calls.

Your device

Recorded as Something else (laptop or desktop) (web_other). Change it if that's wrong.

Where your results go

Results can help build community evidence for follow-up with service providers and regulators. Automatic report emailing is not enabled.

Privacy & location

Tests are saved on this phone first. Real road and spot tests are also uploaded to the project database with precise GPS points and timestamps, the carrier you selected, any notes you typed and a random install ID. If you sign in to an optional account, your tests are linked to it so you can see your own results. Public community pages show only grouped summaries, not routes. Uploaded tests do not currently expire automatically; to ask for help with your data, use the Feedback page. Feedback you send carries no location or route.

Install ID: not created yet

Version

• App: v1-beta.4 (beta-1/report-3)
• Method: beta-1
• Report format: report-3