The RMM would need locations under Clients. Exemple if I have a client called 360 then I would have multiple location like 360 - Office - Server or 360 - Office - Workstation, 360 - Toronto - Workstations. Then your task can simply be executed on the location and you don't do the mistake of running scripts on server. The can be also used to have policies attached like updates, agents etc. that can differ from workstation to servers. The tags are not enough since they there is no visual representation and its easy to make a mistake. The locations are also good for keeping tracks of computer and inventory. I like to create one for SPARES to know what is available. ex: 360 - office - SPARE. In the current state to mimik that I would have to create Clients for each location which is going to do a big mess when you have 50 clients.