Productivity Tracking for Knowledge Workers: What to Log, What to Ignore

You tracked time-in-chair for a month. The total was respectable, the trend line was flat, and the numbers told you nothing you could act on. Some weeks you felt like you got real work done. Some weeks you stared at the screen for nine hours and shipped almost nothing. Your tracker did not distinguish between the two.

If your job is knowledge work, the variable that matters is focus quality, not hours logged. Most productivity tools optimise for the things that are easy to measure, which usually means the things that do not predict your output. This article is about what to track instead, what to ignore, and what the patterns look like once you have four to six weeks of honest data. It is written for developers, designers, writers, founders, consultants, analysts.

What to track

Four fields, plus one optional. None of them are exotic. The point is the combination.

Focus, on a 1 to 7 scale

The headline metric. The single number that, more than any other, tells you how the day went for your actual job.

Set up a scale field, 1 to 7, with half-steps if you want them but whole numbers are cleaner. Log it twice a day. Once at midday, once at end-of-day. Two reads beat one because the day usually has a shape and a single reading flattens it.

Write two or three sentences for what a 1 and a 7 feel like before you start. A 7 is a session where you noticed nothing else for two hours. A 1 is a session where you closed the laptop and could not name a single thing you had moved forward. Without these anchors, a 5 in week one and a 5 in week six will not be comparable.

Energy, on a 1 to 7 scale

The input that most reliably explains focus. If focus is the output, energy is the upstream variable that shapes it. You will see the relationship within a few weeks.

Same scale, same shape. Log it at the same two times as focus. Anchoring the two scales to the same moments makes the pair analysable later.

We covered the mechanics of energy logging in depth in how to track energy throughout the day. Two readings is the minimum that captures the within-day swing. Three is fine. Four starts to feel like a chore.

Sleep, in hours

The strongest day-after predictor of focus for most people. Last night’s hours sit in one row, this morning’s focus sits in the next, and a comparison across a few weeks tends to show a clean gap.

Set up a number field, unit “hours”. Log it in the morning, after you wake up, when your sleep for that night is over and you can put an honest number on it. “About 7 hours, woke twice briefly” rounds to 7. Don’t overthink it.

If hours feels too clinical, a 1 to 7 scale for sleep quality also works. The relationship with next-day focus shows up either way. We unpacked this specific pair in how sleep affects focus.

Deep-work hours, as a number

The output measure that actually correlates with feeling productive at the end of the day.

A deep-work hour is one hour, uninterrupted, single-task, on the work that requires your highest cognitive load. Coding a hard module. Writing a difficult section. Drafting the proposal you have been avoiding. It is not “hours in the editor” or “hours not in Slack.” It is single-task time on the work that mattered.

Log a number at end-of-day. Most days will be between 0 and 4. A 6-hour day is rare and tends to leave you flat by 6pm. A 0-hour day happens more often than you think and is worth knowing about. The number is a slight self-estimate. That is fine: the trend across weeks is what matters.

Optional: caffeine cups, as a number

The most-discussed variable in knowledge-worker productivity, usually less impactful than people assume. If you track it for a month, the connection with focus is often weaker than the connection with sleep or energy. Worth logging if you suspect a real relationship; skippable if your intake is steady.

Five fields total, four required. Two minutes a day, less once it is habit.

What not to track

The honest filter. These are things knowledge-worker productivity tools love and your output mostly ignores.

• Total work hours. Noisy and uncorrelated with output once you are past about 4 to 5 hours of real work. Tracking it incentivises stretching the day, not improving it.
• Meeting count. Useful as context (a Wednesday with 6 meetings is structurally different from one with 1), but not as a metric to optimise. The goal is not “fewer meetings.” The goal is “enough deep-work hours despite the meetings.”
• Slack messages sent. Vanity. Replying fast is not the job.
• Tasks completed. Incentivises ticket-shrinking. Two tickets that move the product forward beat ten tickets that move nothing.
• Pomodoros. A method, not a measurement. If pomodoros help you focus, run them. Logging the count tells you about the technique, not the work.
• Lines of code, words written, files saved. Surface area, not depth.

If you are tempted to track any of these because they are easy to measure, that is the warning. Easy-to-measure is not the same as useful.

The patterns you will see in four to six weeks

Consistent logging across a month produces three or four patterns that are not subtle, plus a few that are. The interesting ones, in rough order of how reliably they show up.

Your focus-curve across the day

The shape of your focus from morning to evening. Most people have a peak that runs two or three hours, a trough somewhere, and a recovery or a fade. The shape is yours and it is more stable than you would guess. After three weeks of twice-daily focus logs, your midday and end-of-day averages will start to settle, and the gap between them tells you most of the story.

This is the same pattern explored in detail in how to track energy throughout the day. Energy and focus tend to share a curve, with focus slightly more sensitive to interruptions and slightly less sensitive to the absolute time of day.

The sleep to next-day focus relationship

The cleanest day-after pair in personal data. Days following 7 or more hours tend to show a focus rating 1 to 2 points higher than days following under 6 hours. The gap is usually wider on weekdays than weekends, because weekend focus is shaped by entirely different things. We covered the specifics in how sleep affects focus.

The reason this pair matters for knowledge workers specifically: focus is your bottleneck, and sleep is the most modifiable input that meaningfully shifts it.

Day-of-week patterns

Monday focus and Thursday focus are often different shapes. Monday tends to be lower for a lot of people, partly because of weekend recovery patterns, partly because the week’s structure has not crystallised yet. Friday afternoons can run high (work pressure releasing) or low (week’s deficit catching up), and which one you are tells you something useful about your own pace.

After six weeks, average your focus by day-of-week and look at the chart. The differences are usually 0.5 to 1.5 points on a 7-point scale, which is enough to plan around.

The deep-work-hours to end-of-day energy relationship

Usually inverse. A 6-hour deep-work day leaves you flat. A 1-hour deep-work day leaves you fine at 6pm and sometimes vaguely unsatisfied. The pattern is not “more deep work is bad.” It is “deep work is expensive, and being expensive is part of why it works.”

Knowing the cost helps you plan recovery. A heavy Tuesday usually means a lighter Wednesday. Pretending otherwise burns through the month.

Seasonal and structural shifts

These show up later. Three months in, you might notice that your November focus baseline is lower than June, or that the weeks after a major launch run lower on energy. These are not breakdowns of the pattern; they are the pattern reflecting that your life has more inputs than five fields.

What to do with the data

Patterns do not make decisions for you. They change the quality of the decisions you make.

Protect your high-focus hours

If your data shows a consistent peak between 9 and 12, defend it. Push status meetings to the afternoon. Decline the 10am call when you can. The peak is not a personality trait; it is a finite resource that shows up in your numbers.

Stop fighting your low ones

If 3pm consistently reads a 3, you can either keep fighting it with coffee and willpower, or you can move the hard cognitive work earlier and accept that 3pm is for shallower tasks. Most weeks, accepting is cheaper than fighting.

Stop attributing bad days to coffee or willpower without checking sleep first

The single most common mistake. You had a low-focus day, you blame the second coffee or the bad night’s lasagna or your lack of grit. Your data, four weeks in, will repeatedly point at sleep instead. Check the obvious input before reaching for the exotic one.

Renegotiate calendar density based on real data

“I have too many meetings” is a feeling. “I have averaged 0.8 deep-work hours on Tuesdays for the last six weeks, versus 3.1 on Thursdays, and Tuesdays have 5 meetings while Thursdays have 1” is a case. The data is not the argument. It is the evidence behind the argument you were going to make anyway.

When you notice a sustained dip, ask what changed

A two-week dip in focus or energy is not the system breaking. It is the system reflecting an input. New project, new manager, a noisy neighbour, a kid sleeping badly, a season change. The data is asking you to look outward, not to add a productivity app.

The freelancer and founder special case

If you are self-employed, more autonomy means tracking matters more, not less.

You are the only person who will notice your patterns. Nobody is monitoring your output, your meeting density, your Thursday slumps. If you do not write the data down, the patterns live entirely as impressions, and impressions drift.

Your decisions also have higher leverage. You set your own calendar. A well-known focus-curve is worth more to a founder choosing what to do on Tuesday than to an employee who does not control Tuesday in the first place. The cost is two minutes a day; the return is decisions made with a clearer view of your own throughput.

Manager conversations and 1:1s

If you have a manager, the data also makes certain conversations easier.

“I’m at my best between 9 and 12” is an opinion. “I’m at my best between 9 and 12, here’s six weeks of focus data showing my morning average is 5.8 and my afternoon average is 4.2” is a position. The same goes for asking to move recurring meetings, requesting blocked focus time, or explaining why a specific project is going slower than the previous one.

You do not have to share the data. The fact that you have it changes how you talk about your own work. It moves the conversation from preference to observation, and good managers tend to take observation more seriously.

FAQ

Should I track project-specific output, like hours on Project A versus Project B?

Sometimes, and coarse categories beat fine-grained ones. A categorical field with three to five options (“client work / internal / writing / other”) logged once a day for which got the most of your focus is enough. Project-by-project timesheets collapse within two weeks because the logging itself becomes the work.

What about screen-time apps that track where my hours go?

Different tool, different job. Screen-time apps tell you where your time went. Loggr tells you how it felt and what you got done. Both are useful; neither replaces the other. Screen time on its own is the same trap as time-in-chair: a count without a quality dimension.

Should I share this with my manager?

Your call. The data is yours. Some managers love it, some are uncomfortable with it, some pretend to care but actually want to know about the project, not your numbers. A safe pattern: keep the raw data private, share the conclusions when they support a request you are making.

How long before I trust the patterns?

Four weeks for the daily metrics, two to three months for trends. Week one is calibration; your scale will drift while you figure out what a 5 and an 8 feel like. By week four, the within-day swing and the sleep relationship are usually visible. By month three, the day-of-week and seasonal patterns are easier to read.

What if my focus barely varies day to day?

Two plausible explanations. Your work is structured tightly enough that focus is mostly determined by external factors, and the day-to-day variation is genuinely small. Or your logging is drifting and the same number is masking different realities. The first is fine; the second is fixable by writing better anchors for the scale.

Do I need to track caffeine or food?

No. The four-field core is enough to start. Caffeine is worth adding if you suspect a real effect and want to test it. Food is data-heavy and easy to abandon; a low-cost version is a categorical lunch field (light, normal, heavy), added only after the first four fields are stable.

Key takeaways

• Time-in-chair is the wrong number for knowledge work. Track focus, energy, sleep, and deep-work hours instead.
• Focus on a 1 to 7 scale, logged midday and end-of-day, is the headline metric. Energy on the same scale at the same times is the strongest upstream input.
• Sleep in hours is the strongest day-after predictor of focus for most people. Deep-work hours is the output measure that actually correlates with feeling productive.
• Ignore total work hours, meeting count, Slack volume, tasks completed, and pomodoro count. Easy to measure is not the same as useful.
• Four to six weeks of logging surfaces your focus-curve across the day, the sleep-to-next-day-focus relationship, day-of-week patterns, and the cost of heavy deep-work days.
• Protect your peak hours, stop fighting your low ones, check sleep before blaming caffeine or willpower, and renegotiate calendar density based on the actual numbers.
• For freelancers and founders, tracking matters more, not less. Nobody else is going to notice your patterns for you.

Try it tomorrow

The smallest useful version: tomorrow morning, open Loggr and add three fields. Focus on a 1 to 7 scale. Energy on a 1 to 7 scale. Sleep in hours. Set a reminder for noon and one for 6pm. Log for four weeks without changing anything else.

At the end of week four, open the day-of-week view and look at the focus chart. That is where the patterns will start to be obvious, and that is where the decisions you make about your week stop being guesses.