Loading
Because anyone can invent one. Within a year there are four spellings of the same tag, none applied consistently, and filtering by any of them returns a third of the answer.
A label is a cross-cutting classification on a task — a theme, a risk, a client concern — independent of which project it sits in.
Labels are defined at workspace level with an owner, so the set stays small and spelled one way. Applying them is open; creating them is not.
A client integration touches four projects. Labelling the tasks lets one filter show every piece of work about it, with hours and status, across all four.
They will not substitute for structure. If a label is really a project, a phase or a status, it should be one of those instead.
Filters, saved views and reporting all read them, and a label can drive a saved view that a team uses as its working queue.
Whoever you nominate. Open creation is what kills tag systems.
Yes, and colour is per label rather than per person.
Yes, per project, where classification actually matters.
Fourteen days, every module, no card. Or half an hour with someone who will run it on your own records and tell you where it does not help.