What a Working Arrangement Looks Like

Rather than a summary, a description of the arrangement these notes point toward.

The purpose

Stated, from the four, in writing, and repeatable by everyone involved.

Recording, billing and costing served from one record of duration.

Monitoring not bundled in. No screenshots, no activity percentages, no automatic idle deduction.

A published purpose limitation that has been tested by a request and held.

The record

Start, end, breaks, recorded to the minute.

Breaks recorded, not deducted.

Rounding absent, or neutral and measured.

Every edit logged and visible to the worker.

Accessible to the worker routinely, without a request.

Retained per jurisdiction, deleted on schedule, verified.

The capture

Recording delay measured, with most entries made within a working day.

Categories few enough to hold in the head, filtered to the person's own work.

Two dimensions: what it was for, and what kind of work it was.

Pre-filled from the calendar where the work is meeting-shaped.

Recording takes minutes per week, and that is measured rather than assumed.

The analysis

Effort by project and by activity type, including unbilled.

Estimate against actual, by category, with the spread.

Realisation where billing applies, with write-offs analysed by cause.

Capacity derived from actuals, not from contracted hours.

Hours beyond contracted, reported in aggregate and acted on.

Nothing ranked by individual.

The governance

Consultation completed, with a written response.

An assessment where anything beyond plain recording is involved.

Rules versioned per jurisdiction, with an owner.

Access by role, logged, audited against stated purposes.

A dispute route with a named person and a response time.

The honesty

Data quality reported on the same page as the analysis: delay, tidiness, catch-all rate, completeness.

The recording method stated, because weekly reconstruction and timer capture are different data.

Known omissions named, particularly unbilled effort.

No precision the method does not support.

Findings reported back to the people who recorded the time, first.

What it produces

Reliable answers to four questions: where effort went, how long work takes against estimate, what was billed against what was recorded, and whether anyone is working more than they should be.

A record that protects the worker and serves as evidence for the employer, which is the same record.

And a workforce that fills it in accurately, because they can see what it is for and can see their own data in it.

None of this requires an unusual budget, and all of it requires deciding what the system is for before deciding what to buy.

Measuring the arrangement itself

Six questions, asked annually, that describe whether it is working.

Can you produce a worker's complete record in minutes?

Is recording delay falling or stable?

Has a finding from the data changed something this year, and was it reported back?

Has the purpose limitation been tested by a request and held?

Can workers see their own records without asking?

Would you design it the same way again?

A no to any of these names the work to do next, which is more useful than a maturity score.

What it costs to run

Ongoing, once established, and smaller than most expect.

An owner with allocated time, which is the real requirement.

A monthly one-page report.

A quarterly data quality check — the four measures.

An annual review of rules, retention, permissions and the purpose limitation.

A category list prune.

None of this is a large commitment, and all of it separates a record people fill in accurately from an administrative chore that produces fiction.

Turn the principle into a test

For an example that can make this requirement testable, consult the ecommerce use case. Treat the page as a starting point rather than proof: reproduce the workflow with real roles, exceptions and permissions before relying on the result.