Know what question you are asking

Before adding any metric, write down the decision it will inform. "Am I doing my deep work in the morning or losing it to meetings?" is a question a focus-time chart can answer. "Am I productive?" is not a question any chart can answer, and tracking against it produces guilt rather than information.

Prefer inputs you control

Output varies for reasons outside your control: some problems are simply harder, some weeks are full of unavoidable interruptions. Input metrics — focus sessions started, days a habit was kept, tasks closed — reflect behaviour you can adjust. Track those, and let outcomes be the slower signal you check monthly.

Look at trends, ignore single days

Daily numbers are mostly noise. A bad Tuesday means you had a bad Tuesday. Four-week rolling views show whether something real is changing. Set your default view to the longest period that still feels actionable — usually a week or a month — and resist checking the daily figure.

Watch for the metric becoming the work

Any number you take seriously will start shaping behaviour, including in ways you did not intend. Counting tasks encourages splitting work into trivially small tasks. Counting focus minutes encourages sitting at the desk without thinking. Notice when you are working to move the number rather than to do the thing, and change the metric when it happens.

A metric is a flashlight, not a scoreboard. It is for finding things, not for winning.

Use a small set, deliberately chosen

One consistency measure — how many days out of seven the habit or session happened.
One attention measure — focused time, honestly logged.
One completion measure — meaningful tasks finished, not micro-tasks.
One subjective measure — a daily mood or energy rating, which often explains the other three.

Four signals is enough to see a pattern. Twenty produces a dashboard you stop reading.

Include how you felt

Quantitative tracking misses the most important variable: whether the week was sustainable. A month of high output followed by a collapse is not a good month. Logging mood or energy alongside activity is what lets you tell the difference between a productive stretch and an expensive one.

Review, then close the dashboard

Check your statistics once a week, draw one conclusion, and change one thing. Then leave them alone. Continuous monitoring adds no information — the data does not change faster than your behaviour does — and it converts a useful tool into a source of low-grade pressure.

Signs your tracking has gone wrong

You feel guilty looking at the app rather than informed.
You have changed nothing based on the numbers in the last month.
You are logging activity in order to log it.
A missed day ruins your mood for the rest of the day.

If two or more of those are true, simplify: keep one metric, review it monthly, and let the rest go.