Semaphore signalling works in the same way hitting an analogue TV used to. It keeps trains moving, but it's a ridiculously inefficient use of capacity and drives up costs as you not only need enough signallers to cover every 'box but S&T need to maintain an entire (completely outdated) skillset to be able to keep the wires and rods in place. That's before we get on to the operational benefit of being able to see a bigger picture in the ROC.
There's a good reason almost every evolution of signalling around the world has ended up with more track controlled from a single place.
Because the driver for signalling replacement has tended to be a need for increased capacity, most Sempahores now left are on the most lightly used lines - Highland Line, S&C, Cumbrian Coast, Welsh Marshes, Heart of Wessex etc. These are lines where passenger revenue is never going to come close to covering costs, so need to get their operating costs under control. The single biggest expense for any business is staff, so it would make a lot of sense for the most staff-intensive form of signalling to be a high priority for replacement.