Recording Hours Against Jobs

No affiliate links on this site, and nothing here is ranked.

Three different tasks under one name

Billing. Hours become an invoice line, and the requirement is accuracy and a record the customer will accept.

Costing. Hours tell you what a job actually took, which is the only honest input to pricing the next one.

And monitoring. Somebody watching whether people are working. A different task with different consequences, and one where the tool changes the behaviour it measures.

Most products serve all three and are sold on the third, which is why the category is larger and more contentious than the task requires.

What billing actually needs

Start, stop, which job, and a note of what was done.

That is the whole record. A timer helps; a manual entry at the end of the day is adequate and frequently more accurate, because somebody who knows what they did writes it more truthfully than a timer left running through lunch.

Where it stops: approval by somebody else, export into an invoicing system, and multiple people on one job. Each of those is a real reason to buy and none is a reason to buy a platform.

What costing needs, which is different

The same data, in a form you can question.

Hours by job, alongside what the job invoiced and what it cost in materials. The comparison is the point, and most time tracking tools do not hold the other two columns.

Which is why a sheet frequently wins for this purpose — not because it records time better, but because it is the only place the three numbers sit together.

Rounded is fine. Costing does not need precision; it needs consistency, and a rounded record kept every week beats a precise one kept for a month.

What to avoid

Anything that measures activity rather than time. Keystrokes, screenshots, application usage. These measure input device activity, not work, and they have documented effects on the people subject to them.

Anything that requires a timer to be running. People forget, and a system that produces wrong data when somebody forgets produces wrong data.

And anything where the entry burden falls entirely on the person who gets nothing back. That arrangement is not stable, and time tracking is the clearest example of it in any category.

The version that gets used

Weekly, retrospective, rounded to the quarter hour.

Not daily — daily entry is abandoned within a month by everybody except the person who required it.

Not a timer — timers are for people who bill in six-minute increments and are a burden for everybody else.

A note per job as you go, and fifteen minutes on Friday turning notes into hours. It is less accurate and it is the version that still exists in six months, which makes it more accurate in every sense that matters.

What it tells you after three months

Which jobs take longer than quoted, which is the finding that changes prices.

Which work you are quickest at, which is frequently not the work you enjoy most.

And how much of the week is unbilled — quoting, travel, admin — which for most small firms is between a third and a half and is the number that explains why a busy month did not pay.

None of that requires a tool. It requires the record and twenty minutes with it.

Where a proper tool earns its place

Three situations, and they are specific.

Several people on the same job, where somebody has to combine entries and see the total before invoicing.

Billing that must be defensible — where a client will query the hours and a printed export with dates and descriptions settles it.

And a volume that makes manual assembly genuinely slow. Beyond a certain number of jobs and people, the fifteen minutes on Friday becomes two hours, and that is a real reason.

Below those, the sheet is not a compromise. It is the better instrument, because it holds the columns that answer the question you actually have.

The monitoring question, briefly

If the reason for the tool is finding out whether people are working, the tool will not answer it.

It will produce numbers, and people subject to measurement adjust to the measurement — which means the numbers describe the response rather than the work.

Where the concern is genuine, it is a management question, and a record of hours is neither evidence nor a remedy for it.

Where the concern is billing accuracy or capacity planning, those are the narrow needs described above and they do not require observing anybody.

Starting from nothing

Two weeks, five categories, no change in behaviour.

Making, admin, quoting, travel, and the actual billable work — or whatever five categories fit what you do.

Rounded to the quarter hour, noted as you go, totalled on Friday.

Two weeks is enough to surprise most people, and the surprise is usually how much of the week is not the work itself. That number is the reason to keep going, and it arrives before the habit has had time to feel like a burden. A separate outside reference is DocuSign.

The short version

• Three tasks share the name: billing, costing, and monitoring — and products are sold on the third
• Billing needs start, stop, job and a note; a manual entry is frequently more accurate than a timer left running
• Costing needs the same hours alongside what the job invoiced and cost, which is why a sheet usually wins
• Avoid anything measuring activity rather than time, anything depending on a running timer, and anything where the entrant gets nothing back
• The version that survives is weekly, retrospective and rounded — less precise and still in use after six months
• Three months in it shows which jobs overrun, which work you are quick at, and how much of the week is unbilled