Rolling It Out

A time tracking rollout is read as a statement about what the organisation thinks of its people. The sequence determines which statement.

The sequence

One: name the purpose, from the four, in writing.

Two: decide what is not collected, and write that down too.

Three: consult, properly, before anything is procured.

Four: design the category list with the people who will use it.

Five: pilot with a team that volunteers, not with the team that most needs watching.

Six: fix what the pilot found.

Seven: roll out, with the purpose limitation published.

Eight: report something useful back within the first month, which is what establishes that the data is for improving work rather than for watching people.

The first decision that sets the tone

Whether activity monitoring is enabled.

It is a setting, it is frequently on by default, and the workforce will find out within days.

Turning it off and saying so, before anyone asks, is worth more than any amount of subsequent reassurance.

Turning it on quietly costs the accuracy of the whole system, for the reasons in the purpose limitation note.

Piloting

With a team that agrees, because a resentful pilot tells you about resentment rather than about the system.

Long enough to include a normal month: a quarter end, a busy week, a holiday period.

Measuring the things that matter: recording delay, time spent recording, catch-all usage, and what people say.

With a defined criterion for proceeding, written before the pilot, including what result would mean stopping.

And with the pilot data deleted or migrated deliberately, rather than becoming the first month of production history by accident.

What the pilot will find

The category list is wrong. It always is on the first attempt.

Some work does not fit: on-call, travel, interruptions, the thing that spans three projects.

The tool is slow somewhere that matters — the mobile case, the offline case, the meeting case.

People want to see their own data and frequently cannot.

Managers want a report the system does not produce, which is the moment to check whether it is one of the four purposes or the fifth.

Rolling out

By team, not by department, so problems are contained.

With the person who ran the pilot available to the next group.

With the purpose limitation published before the first login, not afterwards.

With training that takes ten minutes, because anything longer signals a system that is too complicated to be filled in accurately.

The first month

Report something back, publicly.

Make it a finding about the work, not about people: where effort actually went, which project cost more than assumed, how much of the week is meetings.

Say what will change as a result.

This single act does more for data quality than any enforcement, because it answers the question everyone is quietly asking about what the data is for.

Reporting something back in month one

The single act that does most for data quality, and it costs an afternoon.

Find one finding about the work — where effort actually went, which project cost more than assumed, how much of the week is meetings.

Report it publicly, to everyone recording time.

Say what will change as a result.

Not a finding about people, which would answer the question everyone is quietly asking in the wrong direction.

This establishes what the data is for more effectively than any policy document, and the effect on recording accuracy is visible within weeks.

Piloting with a willing team

The choice that determines what the pilot actually measures.

Not the team you most want to watch, which produces a pilot about resentment.

A team that agrees, doing normal work, over a full month including a busy week and a quarter end.

Measuring recording delay, time spent recording, catch-all usage and what people say.

With a criterion for stopping written before it starts.

And the pilot data deleted or migrated deliberately, rather than becoming the first month of production history by accident.

Connect policy to configuration

The practical choices behind this note can be compared with this rollout reference. Keep the organisation's written purpose in control of the setup, enable only the data needed for that purpose and review the result with affected users.