When the Old Data Stops Being Relevant
Years of record describing a life you no longer live. How to tell which parts still apply and which are history rather than evidence.
A ten-year record spans several versions of you. Some of it informs the present and some of it describes somebody else, and the distinction is not obvious from the data.
To connect this idea with an actual work log, consult the official page. The useful test is whether the resulting entries remain consistent and understandable during a later review.
What ages out
comparison sets for work you no longer do. How long a task took in 2019 with the tools of 2019 is a poor guide to now.
For a neutral companion reference on long-term reference libraries, see Zotero; it helps separate the choice of format from the conclusions drawn from the record.
Capacity figures from a different life stage. Hours available before children, or before an illness, or in a different job.
Categories built around circumstances that ended.
And anything measuring a skill you have since developed, where the early figures describe a beginner.
What does not age
Seasonality, largely. The calendar is stable even when you are not.
Your own estimation bias, which is persistent and personal.
The shape of your recovery from difficult periods.
And the notes, which are readable as history regardless of whether the numbers still apply.
Deciding a cutoff
For estimating: use the last two or three years, or the last twenty comparable items, whichever is shorter.
For seasonality: use everything.
For capacity and workload: since the last major life change, whenever that was.
Different questions want different windows, and applying one window to all of them produces either stale estimates or too little data.
Marking the transitions
A dated line in the definitions file at each major change: new job, move, illness, change of field.
These become the natural boundaries for any analysis.
Without them you will compare across a transition without noticing, and the resulting finding will describe the transition rather than what you were asking about.
What to do with the aged-out years
Keep them.
Exclude them from analyses where they do not apply, deliberately and with a note.
And read them occasionally as narrative, which is a different use and frequently the more valuable one.
The trap of the long baseline
Ten years of data invites using all of it, because more data feels better.
For questions about now, it is worse: an average over ten years describes nobody in particular, least of all your current situation.
Long records are for comparison across periods, not for averaging over them. That distinction is the main analytical discipline a long record requires.
When most of it has aged out
After a genuine change of field or life stage, the older years may inform almost nothing about the present.
They remain worth keeping as history, and the starting-over note covers the case where a fresh structure is warranted.
Do not conclude the record was wasted. It answered questions at the time, and it will be the thing you read in twenty years.
What to check
Which of your questions should use the last two years, and which should use everything?
Are your major life transitions dated in the record?
When you last estimated from your history, how far back did you go?
And are you averaging across a period that contains a transition?
The point
Long records are for comparison across periods, not for averaging over them.
An average over ten years describes nobody in particular, least of all your current situation.
Worth adding
Date the major transitions — job, move, illness, change of field — because they become the natural boundaries for every analysis you will run..
Additionally
Different questions want different windows: estimating uses the last two or three years, seasonality uses everything, capacity uses the period since the last life change..
Finally
Seasonality, estimation bias and the shape of your recovery do not age out.
comparison sets and capacity figures do.
In summary
A ten-year record spans several versions of you.
Some of it informs the present and some describes somebody else, and the distinction is not visible in the data.
Last word
The practice this points at is unglamorous: a small daily act, a short weekly correction, an annual read, and a written page of conventions beside the data. None of it is difficult and all of it is easy to skip, which is why most records do not reach the point where they start paying.
More in this section
These notes are about keeping a personal record for years rather than for a fortnight: durability beats accuracy, and a rough record kept for five years answers questions a perfect record of one month cannot. Tool guides are included as practical comparisons, while the underlying method remains independent of any single product. Nothing here is legal, tax or medical advice: requirements differ by jurisdiction and situation.