Updated: Mar 26
Geolocation is met with unrealistic expectations. For instance, ordering an Uber or pizza and watching a moving dot on your phone as the car drives into your drive is just a fantasy. The Uber or pizza driver has a smartphone with an accelerometer and gyroscope in it, and that phone or app is using GPS satellite positioning. GPS is a fairly accurate technology, but unfortunately often not accessible for online applications.
The most common form of Geolocation methods is IP based, and this is software that is capable of deducing the geographic position of a device connected to the internet. This IP address can be used to determine the country, city, or zip code for the object’s geographical location. These services only provide 55-80% accuracy for a user’s region or state, and they provide 50-75% accuracy for a user’s city. It’s based on non-cutting edge technology that hinges upon a giant database of IP addresses. If the database is incorrect, then so will the determination of the geographic location. Cell phones obtain new IP addresses as they move, and cell service providers service users across the country. This makes it even more difficult to find a precise location of a specific IP address.
One example of inaccurate IP-based GPS: testing in rural Saskatchewan often would geolocate to Regina, the home office of the government owned telecom and hundreds of miles from the correct location. In the same province as Saskatchewan, Bell mobile phones are geolocated to Calgary. Corporate networks are often located to wherever the head office is located.
With HTML geolocation API, Geolocation is more accurate. This browser based geolocation must receive permission from the user before it can access their location. GPS on smartphones can be accurate up to 95% within 3 meters if 4 satellites are being utilized. When 3 satellites are being used, accuracy is only slightly reduced. It’s still slow and has problems indoors, around buildings, or around mountains. Wi-fi positioning leverages major geolocation providers such as Google, Apple, and Microsoft, who are steadily gathering huge amounts of data on public and private wi-fi endpoints and SSIDs (network identifiers). Looking to mobile network connections, devices are triangulated with 3 cell towers, but this can be inaccurate depending on available towers and the distance between them. The browsers have a hierarchy, defaulting to the most accurate technology first and moving down list until one of the methods is available.
There are many ways to locate people who are online using technology, but there’s not yet a single workable solution across the board. Any and all technology is subject to pain points, especially for SaaS providers.