Ten years ago, specifying a variable frequency drive on a pump set was a decision that needed justifying. This year it is unusual for one of our booster or process sets to leave without one. The technology did not change; the economics did.
Three things converged
Price. Drives have fallen in cost every year while capability rose. The premium that once bought only large installations now buys drive control on domestic-size sets.
Integration. What used to be a cabinet-mounted component is now built into the motor or the pump controller, wired and tuned at the factory. The installation penalty largely disappeared.
The energy case got impossible to ignore. A pump's power scales with the cube of its speed; trimming 20 percent off speed cuts energy nearly in half. In variable-demand duty, which is most real buildings, the drive pays for itself in a couple of years and then keeps paying.
Where fixed speed still makes sense
Steady, predictable loads: irrigation blocks that run flat-out on a schedule, process duty at one operating point. There the drive adds cost and a standby draw without a load to follow. But that is the shrinking minority. Variable speed has become the default not through fashion but through arithmetic, and it is difficult to see the argument reversing.