Tool Migration Without Losing Years
Moving a multi-year record to a new tool. The sequence that preserves history, and the shortcuts that quietly destroy it.
Migration is where long records die, usually not at the moment of the move but months later when somebody notices the old years are gone or wrong.
For a practical application of this principle, see daily work tracking. It provides a supporting reference for testing a small real-world sample without adding unnecessary friction to the record.
Before you move anything
Export from the old tool and verify the file, using the checks in the export note.
For another established perspective on focus intervals, the Pomodoro Technique provides a useful reference beyond any single tracking product.
Keep that export permanently, whatever happens next. It is the only artefact that certainly contains your history in its original shape.
And do not cancel the old account yet. The cost of a month's overlap is small against the cost of discovering a problem with no way back.
The sequence
Import into the new tool with a small sample first — one month — and check it arrived correctly.
Look specifically at: dates at year boundaries, durations, categories, tags, and notes.
Then import everything.
Then run the same total on both systems for a period you know, and compare. If they disagree, stop and find out why before proceeding.
What commonly breaks
Time zones. Entries shift by hours, and the shift is invisible unless you check a known entry.
Category mapping. The new tool has different categories and the import quietly puts everything in one.
Notes. Dropped or truncated.
Entries spanning midnight, which different tools split differently.
And the oldest years, which some importers silently skip beyond a range limit.
When the new tool cannot hold the history
Common, and not a disaster.
Keep the old record as a static archive — the verified export, stored — and start the new tool from the migration date.
Record the date of the break in your definitions file.
A record in two segments with a documented boundary is far better than a merged record with silent corruption, and analysis across the boundary is still possible.
The parallel period
Run both for a week if you can.
It catches the differences in how they define things: what counts as a day boundary, how they round, whether a paused timer counts.
These differences are the source of the discontinuity people find in their charts a year later and cannot explain.
After the move
Note the date and the reason in the definitions file.
Re-run one analysis you have run before and check the historical part is unchanged.
And keep the old export. Permanently. Storage is free and the alternative is trusting an import you verified once.
The shortcut that loses records
Importing everything, glancing at the recent weeks, deleting the old account.
Recent weeks always look fine. The damage is in 2019, and it is found two years later when the account is gone and the export was not kept.
What to check
Do you still have exports from before your last migration?
Did you verify the oldest year after importing, or only the recent months?
Is the migration date recorded anywhere?
And do totals for a pre-migration period still match what they used to be?
The point
Keep the pre-migration export permanently.
The damage from a bad import is in the oldest years and is found two years later, when the old account is gone.
Worth adding
Run both tools in parallel for a week if you can.
It catches differences in day boundaries, rounding and how a paused timer counts — the source of unexplained steps in charts a year later.
Additionally
The shortcut that loses records: import everything, glance at recent weeks, delete the old account.
Recent weeks always look fine and the damage is in the oldest year.
Finally
Do not cancel the old account at migration.
A month of overlap costs little against discovering a problem with no way back.
In summary
Verify with a sample first, then import everything, then run the same total on both systems for a period you know.
If they disagree, stop and find out why.
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.