Glossary and Where to Start

Backfill — adding an entry after the fact. Within a few hours it is a record; beyond a day it is an estimate and should be marked as one.

To connect this idea with an actual work log, consult the feature summary. The useful test is whether the resulting entries remain consistent and understandable during a later review.

Category drift — the same label meaning something different over time. The most damaging failure in long records because it is invisible in the numbers.

A related source worth keeping beside this note is Box, particularly for decisions involving managed file storage.

Coverage — recorded days as a proportion of the period. The figure that makes two periods comparable, and the one most often omitted from analysis.

Definitions file — one page beside the data recording what each category means, what conventions apply, and every dated change to either. The single artefact that makes a ten-year record interpretable.

Gap convention — the agreed meaning of a missing entry, and how it is marked. Without one, a recorded zero and an unrecorded day are indistinguishable.

Granularity — how finely the day is divided. A cost paid daily, and it should be matched to the question rather than to what the tool makes easy.

comparison set — a set of comparable completed items with their actual durations, used for estimating. The highest-value output of a long record.

System rot — gradual degradation while entries continue. Detected by reading entries rather than by looking at totals.

Terms used loosely elsewhere

"Time tracking" covers two different activities: a bounded investigation with a question and an end date, and a continuous record kept for years. This collection is about the second; the requirements differ substantially.

"Accurate" usually means precise. For a long record, consistent matters more than either.

"Complete" is not achievable and not the goal. Marked gaps are fine.

Where to start

Setting up something intended to last: what a long-term system has to survive, durability over accuracy, and what to decide before the first entry.

Struggling to keep it going: boredom, when the habit slips, and logging discipline without perfectionism.

Coming back after a break: recovering a lapsed system.

Changing tools: tool migration, and when to change tools and when not to.

Have years of data and have not read it: the annual read, and questions only long data answers.

Something feels wrong with the record: system rot, category drift, and the records you stop trusting.

If you read only three

Durability over accuracy, because it is the decision everything else follows from.

Missed days, because gaps end more systems than any other single thing.

And what a durable arrangement looks like, because it is the checklist.

A closing note

Nothing here is legal, tax or medical advice. Retention requirements, evidential standards and anything concerning health differ by jurisdiction and by situation, and the specifics come from a professional.

No product names. The useful distinctions are between formats and conventions rather than between tools, and any tool named here would be discontinued before the record was.

What the collection argues

A system kept for years is a different object from one kept for a fortnight, and designing the second and running it as the first is why most long records fail.

Durability beats accuracy: a rough record kept for five years answers questions a perfect record of one month cannot.

Everything is a cost paid daily and a return collected in years, which means the daily act must be small enough to survive indifference.

And the reading sustains the recording. A record nobody consults stops being maintained, in that order.

The point

The reading sustains the recording.

A record nobody consults stops being maintained, in that order, and scheduling the read is the single most effective preventive measure available.

Worth adding

No product names appear here, because any tool named would be discontinued before the record was.

The useful distinctions are between formats and conventions.

Additionally

Time tracking covers two different activities: a bounded investigation with an end date, and a continuous record kept for years.

The requirements differ substantially.

Finally

Complete is not achievable and not the goal.

Marked gaps are fine, and a record that admits them is more usable than one pretending to be whole.

In summary

A system kept for years is a different object from one kept for a fortnight, and designing the second while running it as the first is why most long records fail..