Geofencing draws a virtual boundary around a real place and triggers an action when a device crosses it. The boundary is usually a radius around a coordinate; the trigger is usually a notification, an ad bid, or a change in what content is served.
In a screen network it is the difference between showing one loop everywhere and showing something that knows where it is. A campaign can go live only on screens inside a shopping centre, or a promotion can reach a visitor's phone as they enter the car park while the screens inside carry matching creative.
The caveat is consent, and it is not optional. Device-side geofencing needs location permission the visitor actively granted, and under GDPR and KVKK location is personal data, so "we geofenced everyone who walked past" is a compliance problem, not a campaign. Screen-side targeting carries none of that weight, because the screen's location is fixed and known and nobody is identified.
Most of what is sold as geofencing in signage is really screen-side scheduling, which is simpler, cheaper and safer. Establish which one is being proposed before evaluating the price.