Tracking Wins at Work: The Three-Line Daily System

The reason most people cannot answer 'what have you been working on?' is not that they have not been working — it is that they have not been writing it down. The fix is not a productivity overhaul; it is three lines a day, written at the same time, in the same place, for ninety days. This is the system, what counts as a win, and the templates by role.

The three-line daily entry

Every working day, write three lines. Same time, same place. The lines are: what I shipped today, what I learned today, what is blocking me. That is the entire system. The constraint is the point — three lines is short enough to do on a bad day and long enough to capture a good one.

• Line 1 — Shipped: an artifact, a decision, an outcome. Specific. PR number, doc title, meeting result.
• Line 2 — Learned: something you didn't know yesterday that you know today. Especially: a moment you realized you were wrong.
• Line 3 — Blocked: what is in the way of the next thing. Often the most useful line at year-end because patterns emerge.

What counts as a win

Most people undercount their wins. The default mental model is 'a win is a launch,' which means a person who ships once a quarter logs four wins a year and feels invisible. The wider definition: a win is any verb that produces a durable artifact or outcome.

• Shipped — code, content, a decision document, a feature, a bug fix.
• Wrote — an RFC, a runbook, a memo, a one-pager that someone else used.
• Decided — picked option A over B with reasoning written down. Decisions count as wins.
• Unblocked — got someone else moving who was stuck. Naming who and what and when matters.
• Stabilized — fixed a flaky test, paid down a debt, shortened an oncall pageload. Quiet wins.
• Mentored — paired with someone, ran a review, gave feedback that landed. Compounds in the growth section of reviews.
• Learned — read the source code of a system you depend on, finished a course, came out of a postmortem with a specific lesson.

When to write the entry

End of day, before you close work. Two reasons. First, it is the freshest moment — the artifact is still on screen. Second, the act of writing it ends the day cleanly, which most people are bad at and which is more valuable than the entry itself.

If the end-of-day moment is unreliable for you (managers and PMs especially — your day ends in someone else's meeting), write the entry first thing the next morning instead. The cost is some lost detail; the benefit is consistency. Consistency beats accuracy.

The Friday review ritual

Once a week, fifteen minutes, re-read the week's entries. Pull the two or three biggest wins into a separate weekly summary. Note any patterns in the blocked line — if the same thing has been blocking you for three weeks, it is now a project, not a blocker.

Most people skip the Friday review. The Friday review is what makes the daily entries compound into a brag document later. A daily journal without a weekly review is a heap; with one, it is an index.

Templates by role

Software engineer

• Shipped: PR #4421 merged — fixes the rate-limit edge case in the auth proxy. Backport queued.
• Learned: rediscovered that our retries don't honor the X-RateLimit-Reset header. Going to write up a one-pager.
• Blocked: waiting on platform team to provision the staging keyspace for the migration test.

Product manager

• Shipped: Q2 OKR review draft circulated. Got sign-off on dropping the partner integration goal.
• Learned: pricing tests need a baseline lock at the start of the test, not the end. Found out the hard way.
• Blocked: cannot get a clear answer on what counts as 'activated' for the new dashboard. Will force the question Friday.

Designer

• Shipped: V2 of the empty-state flow — three frames, handoff to eng done.
• Learned: copy-first wireframes get faster engineering buy-in than visual-first. Going to default to copy-first for next two projects.
• Blocked: research review on the onboarding redesign keeps slipping. Will book a recurring 30-min slot.

Manager

• Shipped: closed the staff-engineer hire — start date confirmed for next month. Ran the calibration meeting for H1 ratings.
• Learned: my 1:1s drift to status updates by default; switching to one open-ended question per week as the lead.
• Blocked: still don't have a clean line of sight into team capacity for Q3. Going to write a one-pager on what I'd need.

Sales / customer-facing

• Shipped: closed [Acme] — $42k ACV. Three meetings, two stakeholders. Notes filed in CRM.
• Learned: the procurement objection on data residency is a forcing function — three accounts have raised it now.
• Blocked: waiting on legal on the [Globex] redline. Reminder set for Tuesday.

How wins compound

A single day's three lines look trivial. Ninety days of them, indexed by a Friday review, are something else: a brag document writes itself, a self-review takes ninety minutes instead of three days, and the next-time-someone-asks question has a real answer. The compound effect is also internal — people who track wins consistently report taking on more visible work, asking better questions, and noticing their own patterns earlier.

Common pitfalls

• Writing only on good days — biases the record toward highlight reels and breaks during slumps.
• Adding more lines — three is the constraint that makes the system survive bad weeks.
• Treating it as a to-do list — the entry is what you did, not what you should do.
• Skipping the blocked line because it feels negative — the blocked line is where year-over-year patterns surface.
• Skipping Friday review for a month — the entries stop compounding into anything reusable.
• Trying to make it 'look good' — you are writing for future you, not for an audience.

Tracking wins is the cheapest professional habit you can build and the one that compounds the most. Three lines a day. Same time, same place. A Friday review you skip half the time but do at least twice a month. In ninety days you will have a record no one else on your team has, and the only person who has to keep it is you.