Operability asks how easily and reliably a user can operate the product once they know it. A measure relates successful operations to attempted operations, or effort to the task. Operability is the steady state. It covers the number of steps, the error rate, the consistency of the interaction and the physical effort. The measure should be taken from a task that represents normal work, with users who know the product, because otherwise it measures learning. A common failure is to measure the happy path only, where operability is highest and least informative. The measure should include recovery from an error, which is part of operating the product.
The graph holds this control, the 0 it maps to, and the evidence behind each claim, over MCP and REST.