Why do we need poles? The electric force is the attraction or repulsion between charges. If we for example had a metal with only positive charges, and another metal with only negative charges. The two metal pieces will then attract each other by the electric force. In an electric system, no poles is mentioned. 
The magnetic force is just a relativistic side effect of the electric force, and the difference is that it is created by moving charges and acts on other moving charges. If we have the two metal pieces, they will also feel the magnetic attraction because of the particles inside the metals have motion. The two metals is then said to be magnets. However, in a magnetic system, poles is mentioned.


*

*Why are there only said to be poles in magnetic systems, and not electric systems.

*Why does a magnet always have two magnetic poles? And what is the point of introducing the concept of poles on the metal pieces mentioned above?

*Does poles really "exist"? Are they real concepts, or are they just a way of visualising different properties?
 A: *

*We $do$ speak of electric dipoles. For example, certain molecules give rise to an electric field outside them approximating to that dipole, a pair of equal and opposite, very close together charges. [This can be the first step in explaining why a dielectric between the plates of a capacitor increases the capacitance.]

*A current-carrying loop, at distances large compared with the loop diameter, gives rise to a magnetic field that varies in an exactly similar way in magnitude and direction with position relative to the loop, as the electric field from an electric dipole! Therefore one can think of the loop replaced by two (close together) equal and opposite magnetic poles. This works pretty well for visualising a the field due to a bar magnet, or current-carrying coil, some way away from them. It's even not bad as a first approximation for the Earth's magnetic field. However, applied to what's going on inside magnetic materials, it can easily lead the unwary astray. One has to correct for an important difference between a loop's field (or the field due to the spin of an electron) and a true dipole field: the loop's field lines are continuous unidirectional closed paths linked with the current loop, whereas a true dipole's field lines would start on one pole and finish on the other (so if we insisted on thinking of the pattern as made of continuous paths, part of each path would have to be running backwards). 

*For reasons I've tried to explain, I'd lean towards 'not real', but there are philosophical issues here which I'm not competent to comment on.
