The Annual Read

Most people with long records never read them properly. They add daily, glance occasionally, and form their view of the year from the most recent difficult month.

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

When

A fixed date. The turn of the calendar year, the end of your financial year, or an anniversary of the record.

The broader practice of spreadsheet history is also covered by Google Sheets help, which is useful when designing a record that should remain understandable later.

The same date each time, so the comparison is like for like.

In the diary a year ahead, because this is exactly the sort of task that never becomes urgent.

What to have ready

The year's totals by category.

The same figures for the previous two years.

A month-by-month chart with one line per year.

Your definitions file.

And the notes column, filtered to non-empty. That last one is what makes the hour worth having.

The sequence

Read the notes first. Before any numbers: what the year actually contained, in your own words at the time.

Then the monthly shape against previous years: where did they agree, where did this one depart.

Then the totals, with the coverage figure beside them.

Then the one measure you follow, across all years you have.

Numbers last. The narrative first. People who do it the other way round interpret the year to fit whatever the first chart suggested.

The four questions

What was unusual about this year, and does the record show why?

What has changed across three years — direction, not level?

Is the record still answering a question I have?

And what should change: in my work, or in how I record it?

Writing it down

Half a page, dated, stored beside the data.

Next year you will read it before starting, and the comparison between what you concluded then and what the record shows now is frequently the most informative thing available.

People who keep these annual notes for five years have something genuinely unusual: a record of their own interpretations as well as their own activity.

What to change afterwards

At most one thing about the recording.

Categories that no longer fit, a field that has gone unfilled all year, granularity that has become too fine.

Date the change in the definitions file. The comparing note explains why.

What not to do

Do not restructure everything because the year was difficult.

Do not delete the years that look bad.

And do not conclude a trend from one year, which the trends note covers.

If the read produces nothing

That happens, and it means the year was ordinary — which is itself worth knowing and is not a failure of the record.

It may also mean the record no longer answers a live question, in which case the honest options are a new question or stopping, and both have their own notes.

What to check

Have you ever read a full year of your own record?

Is a date in the diary for the next one?

Do you have last year's written conclusion to compare against?

And when you last formed a view about your year, was it from the record or from the last six weeks?

The point

Read the notes first, then the shape, then the totals.

People who start with a chart interpret the year to fit whatever the chart suggested.

Worth adding

Write half a page after the annual read, dated, beside the data.

Comparing what you concluded last year against what the record shows now is frequently the most informative thing available.

Additionally

Change at most one thing about the recording after an annual read, and date it.

Restructuring everything because the year was difficult is how records break.

Finally

Read the notes before any numbers.

People who start with a chart interpret the year to fit whatever it suggested.

In summary

The four questions: what was unusual, what has changed across three years, is the record still answering a live question, and what should change..

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.