Nope, this is correct. Each data point is clustered based on feature similarity (ie. clustering employees based on their characteristics/features). In theory, you wouldn’t be able to obtain more clusters as the number of features but at that point clustering if pointless as you haven’t really identified any meaningful groups. Hope that helps