It matters to a smaller MSP mainly as a source of proven shapes. The distinction between an incident (restore service now) and a problem (stop it recurring) is worth adopting whether or not you adopt anything else, because conflating them is how the same outage happens quarterly.
The common mistake is implementing it wholesale. It was written with large enterprise IT departments in mind, and a twelve-person MSP applying it in full creates process weight that slows the thing it was meant to improve. Take the vocabulary and the distinctions; leave the ceremony.