There is a lot of debate on this actually, but the consensus is that at best, a soft starter does not significantly add to the motor heating during start-up, as long as it is a successful start. The added acceleration time is somewhat offset by the lower current and vice versa. That of course is the opposite of what you wanted to hear, sorry to burst your bubble. Adding a soft start will not increase the starts-per-hour capability of a motor, but if properly applied will not likely decrease it either (at least in my opinion, but there are many people who disagree). You should also note that each Decel ramp must be counted the same as a start cycle in the starts-per-hour capability of the motor, such as in the case of a pump application.
The type of soft starter also makes a difference by the way. Many newer-smaller-cheaper versions are being made now with what is called "2-phase control" which means they have SCRs on only 2 of the 3 phases, the middle phase is a piece of bus bar. This style, while not unsuccessful, does increase the motor heating during starting because you will have severely unbalanced current in the earliest stages. Unbalanced current creates negative sequence currents in the motor which disproportionately heats the windings above what current flow alone would do. My rule-of-thumb on these is to cut the starts-per-hour capability of the motor in half when using this kind of soft starter. I have no mathematical basis for it, it's just my conservative nature.