Actually, there is one simple way to make it slightly more interesting (at least by giving a real tactical, if limited choice):
Keep the building new seats feature/pricing as is. Introduce a feature to change seat types, where the change operation costs the difference of the two seat types (or slightly less), which takes as much time as building that amount of seats from scratch, and where those seats are out of use for the duration of the work. This way you can balance some small cost saving vs. lost income from seats that cannot be used.