Measuring the System, Not the Person
The same data arranged two ways answers different questions. One produces improvement and the other produces defensiveness.
Measurement can follow the work or follow the worker. The choice determines what the figures can be used for and how people respond to them.
The process issue in “Measuring the System, Not the Person” is easier to diagnose when effort can be connected to projects without confusing activity with value. Used for how employee monitoring works, Monitask can add time and workflow context, while the process outcome remains the source of truth for whether the system improved.
Following the worker
Output per person, activity per person, utilisation per person.
For an independent perspective related to “Measuring the System, Not the Person”, consult the Microsoft WorkLab; it offers a useful external check on definitions, governance and the assumptions built into a proposed measure.
Answers: who produced what.
Produces: comparison, ranking, explanation of individual variation that is mostly caused by something else.
Following the work
How long items take, where they wait, how often they come back, where they get stuck.
Answers: what is happening to the work.
Produces: a list of places to intervene, none of which is a person.
Why the second works better
Causes live in the process more often than in the people, which the systems note argues.
Nobody is defending themselves, so the information stays accurate.
And the output is actionable: a queue can be removed, a handoff eliminated, an approval delegated.
The same data, two arrangements
A ticket system records who and what.
Arranged by person, it is a performance report.
Arranged by item journey, it is a process map.
The second is almost always more useful and takes the same data, which is worth knowing before anybody buys anything new.
What it looks like in practice
Lead time by stage.
Queue length and age at each handoff.
Rework rate by cause.
Flow efficiency: working time as a share of elapsed time.
Four measures, none of which names anybody, all of which point somewhere.
The objection
"But some people are slower."
True, and that is a management conversation based on observation, not a measurement programme.
The error is building an organisation-wide measurement system to detect a handful of cases that a competent manager already knows about.
The cultural effect
Process measurement is discussed openly because nobody is exposed.
Teams will tell you where the problems are, in detail, when the figures are not about them.
That candour is the most valuable output and it disappears the moment measurement becomes personal.
Where individual data is genuinely needed
Its own note covers this, and the list is short: billing, defined regulatory requirements, and a specific concern handled as a conversation rather than as a report.
What to check
Is your measurement arranged by person or by item?
Could you produce a lead-time-by-stage view from data you already have?
Do teams talk openly about where work gets stuck?
And how many of your measures name individuals?