Get your current coordinates, and see how fast your phone finds its position, how accurate the fix gets, and whether the GPS is actually being used. Nothing leaves your device.
Your browser will ask for your location. It stays on this page: nothing is sent, stored or drawn on a map.
Take the device outside, away from tall buildings, and hold it still. Press Start and leave the test running for about a minute — the first fix is quick, the accuracy keeps tightening after it.
The page asks the browser for the most accurate position it can get, keeps asking for as long as you let it, and measures what comes back: how long the first answer took, how tight it got, and how much it wanders while the device stays put.
Press Start, allow location access, and your coordinates appear as soon as the first fix arrives, updating with every new one. They are given in two forms: decimal degrees, like 48.858370, 2.294481, which any map or navigation app accepts in its search box, and degrees, minutes and seconds, like 48°51′30.1″ N 2°17′40.1″ E, the form used on paper maps and in aviation. Both are on WGS84, the same datum GPS itself uses. Six decimal places is about ten centimetres, finer than any phone can actually place itself, so how far to trust the last digits depends on the accuracy shown above them.
Every position the browser reports comes with a radius: the device is 95% sure it is somewhere inside that circle. A phone with a working GPS under open sky gets it down to three to five metres, and between tall buildings or under trees it settles at ten to thirty. That radius is the browser's own estimate, not a measurement against the truth, which is why the page also plots where the fixes actually land — a receiver that claims four metres while its dots wander over forty is one to be suspicious of.
A receiver that has been off for a while has to find the satellites from scratch and download where they are, and that takes from a few seconds to a minute. Phones shortcut it with data from the network, so a first fix in under ten seconds is normal outdoors, and under thirty is fine. A phone that takes minutes in the open, every time, is one whose antenna is hearing the satellites badly.
The browser does not say where a position came from. It may be a satellite fix, a guess from the wifi networks in range, or the area of a cell tower or an IP address. The page reads it from the numbers instead: satellite fixes get under ten metres and usually carry a height, wifi ones sit at twenty to a hundred metres without one, and anything in kilometres is an area, not a position. A laptop has no GPS, so on a desktop this test measures its wifi positioning, and says so.
A browser does not see the satellites. Native apps can list each one with its signal strength, and that is the test to reach for when a satellite-by-satellite view is what you need. What a page can measure is the result — the position, its accuracy, how fast it arrives — and for the question of whether a GPS works, that is enough.
On most phones the GPS antenna is a strip of metal in the frame, or a flex cable to the mainboard, and a back glass or housing replacement is the usual way it gets disconnected or damaged. The symptom is a phone that finds its position indoors on wifi exactly as before, and never gets better outside. Run this test outdoors before and after the repair, and compare the typical accuracy and the time to reach twenty metres.
Your position stays on this page. It is not sent to our server, not stored, and not drawn on a map — map tiles are fetched by location, so drawing one would tell whoever serves them where you are. Your coordinates only leave the page if you choose Open in Maps, and Blur hides them before a screenshot. Analytics receive only how long the test ran and how good the result was.