Reading the numbers · 2.6
What no time record will tell you
What no time record will tell you. What decides it, what it costs, and what usually goes wrong. For a practical comparison point, see Monitask's nonprofit time tracking software.
Six questions people ask of time data that it cannot answerThe limits
Was the work any good? An hour recorded against a task says nothing about what came out of it.
Should this have taken this long? Requires knowing what the work involved, which the record does not contain. For related guidance and industry context, see Atlassian.
Was this hour worth paying for? A judgement about value, made by somebody who understands the work.
Is this person struggling? Sometimes visible as a pattern and never established by one. The answer comes from a conversation.
Why did this project overrun? The record shows where hours went, not why.
Who is our best person? Not a question a time record has any business being asked.
The prevention paradox
Work that stops problems happening produces no events. The engineer whose systems never fail, the account manager who catches a misunderstanding early, the supervisor who notices a fault before it becomes a stoppage: each has produced an absence.
Absences appear in no report. Meanwhile whoever allowed a failure and then resolved it heroically has produced visible hours against a visible problem.
Any system of measurement that does not account for this rewards firefighting over prevention, and almost none does.
Effort is invisible
Two hours of straightforward work and two hours of difficult work are the same two hours. Nothing in a duration distinguishes them, and the person who takes longer may be doing the harder part.
This is the specific reason comparing individuals on time data is not merely unwise but meaningless: the denominators differ and are unobservable.
Correlation, and the trap in it
Time data produces correlations readily and explanations rarely. Clients with more hours are less profitable; teams recording more internal time deliver faster; people who log fewer hours leave sooner.
Each of those could run in either direction or be caused by something else entirely. Treating them as findings is how organisations arrive at confident policies built on nothing.
What to do about all this
Use the record for what it evidences: where hours went, and the claims in the first entry of this part. Use conversation for everything else, and be explicit with the team about which is which.
An organisation that says out loud what its data does not measure gets better data, because people stop guarding against uses that have been ruled out.
What the record does establish
Worth restating after a page of limits: where hours went, that they were recorded at the time, by whom, against what, with corrections visible.
That is a narrow claim and it is enough to support an invoice, a payroll run, a capitalisation split, a capacity case and an audit response. Most organisations need nothing more and try to extract considerably more.
The temptation as data accumulates
Three years in there is a great deal of it, and somebody will propose analysis the record cannot support. The limits in this entry do not weaken with volume: a million rows about duration still say nothing about quality.
Talking to people
Everything this entry lists as unanswerable is answerable by asking somebody who knows the work. That is slower, it does not produce a chart, and it is the only route to the questions that matter most.
Two more the record cannot answer
Whether somebody is overloaded. A full week and an overwhelming week look identical. The difference is in what the person can absorb, which nothing here measures.
Whether a team is well organised. Hours distributed evenly may mean smooth operation or may mean everybody is equally stuck.
The honest use of this part
Read the six entries as a set of constraints on what to claim. An organisation that respects them gets a record it can rely on for a narrow set of important things.
One that ignores them gets a large dataset supporting confident conclusions about matters it never measured, which is worse than having no data at all because it is persuasive.
Where to start
Write down the three questions your organisation most wants answered, and mark which of them a time record can address. The ones it cannot are the ones to stop expecting it to, and to answer some other way.
The most useful sentence in this part
Knowing what an instrument does not measure is the difference between using it and being used by it. Everything else here is an elaboration of that.
What comes next
The next part covers the legal position, which is the other constraint on what a record can be used for, and it differs by jurisdiction more than anybody expects.
The conversation that replaces the analysis
Fifteen minutes with somebody about how their month went produces more of what an organisation actually needs than any report built from durations, and the report is frequently commissioned because the conversation feels harder to justify.
It is also the only way to learn about the prevention work described above, because the person doing it is the only one who knows it happened.
A short checklist
Use the record for the six claims in the first entry. Do not ask it about quality, value, effort or intent. Read correlations as questions. And say out loud what it does not measure.
One sentence to carry
A time record answers where hours went. Everything else asked of it is inference, and the inference is where organisations go wrong.
Also in reading the numbers
Why timesheets are left blank
A timesheet asks somebody to reconstruct a fragmented day hours later. The reconstruction is a guess and everybody knows it.
Choosing the categories
The list somebody picks in week one governs everything the record can later answer.
Reminders, nudges and escalation
Reminders, nudges and escalation. What decides it, what it costs, and what usually goes wrong.
What a manager should do with it
What a manager should do with it. What decides it, what it costs, and what usually goes wrong.