Gross Effective Hours

Gross hours count presence or paid time. Effective hours count the part of that spent on the work itself.

Everything between them is real and mostly necessary: breaks, handovers, setup and shutdown, waiting for inputs, travel between tasks, meetings, system downtime, and the interruptions that come with any role where other people need things.

• Breaks, including the ones statute or the applicable enactment requires.
• Setup, changeover and cleandown in physical work.
• Waiting time, where the employee is available and the work is not.
• Coordination: meetings, handovers, approvals and the queue for a decision.
• Recovery from interruption, which costs more than the interruption itself.

None of this is waste by definition. A support role that is fully utilised has no capacity to respond, which is the opposite of what the role is for.

Comparing utilisation across these categories produces a ranking of job types rather than of effort, which is the most common misreading of the measure.

It also creates a perverse incentive where it is used as a target. A support team told to raise utilisation can do so by responding more slowly, since a queue raises utilisation and degrades the service.

The measure earns its place in capacity planning, where the question is how much work a given headcount can absorb.

Planning on gross hours overstates capacity, and consistently. A team of ten on forty gross hours does not have four hundred hours of work in it, and plans built as though it does produce overruns that get attributed to individual performance rather than to the arithmetic.

It is also useful for finding structural waste rather than individual slack: a changeover that takes longer than it should, an approval step that leaves people waiting, a system that is unavailable at predictable times, or a meeting pattern that fragments the day past the point of usefulness.

The productive question is never why is this person at seventy per cent. It is what happens in the other thirty per cent, and how much of it is avoidable.

The measure turns harmful when it is applied to individuals as a performance standard.

Doing so mostly identifies who has the least controllable workload: the person covering interruptions for the team, the one who trains new joiners, the one whose work depends on inputs arriving. Those are frequently the people an organisation can least afford to discourage.

It also drives recording behaviour rather than working behaviour. Where a number determines how someone is judged, the number is what gets managed, and the recorded figure detaches from reality without anyone deciding to be dishonest.

Surveillance-based capture, whether keystroke monitoring, screenshots or activity tracking, makes this worse rather than better. It produces data that measures visible activity rather than contribution, is widely resented, and gives people a strong reason to generate activity. The data that results is not more accurate for having been collected involuntarily.

• Measure at team and process level, not at individual level, unless there is a specific reason that survives being said out loud.
• Set expectations by role type rather than applying one utilisation target across different kinds of work.
• Treat a low figure as a question about the process before it is a question about the person.
• Explain what the measure is for and what it will not be used for, and then honour that, since one contrary use ends the reliability of the data permanently.
• Prefer self-recorded time in roles where people have reason to record honestly, and invest in the reason rather than in the monitoring.

The fourth point is the one that determines whether any of this works. Time data is only as good as the willingness to record it accurately, and that willingness is destroyed quickly and recovered slowly.