MELLORCATalk to Mellorca
Moments

Time & Attention

The Report Someone Rebuilds by Hand Every Friday

Every Friday afternoon one person exports, pastes, fixes and formats the same weekly report. What that ritual takes from their week and their attention, and what changes when the report builds itself.

Mellorca Moments2026-10-04

At about two o'clock on a Friday, one person in the business opens the same five things. The sales system, to export the week's invoices. The stock system, for movements. The bank, for receipts. The job tracker, for open work. And last week's spreadsheet, which will become this week's report once everything has been pasted in, fixed and checked.

The columns never come out quite the same way twice. A customer name has a trailing space. Dates arrive as text. A formula that worked last week now points at the wrong row. By the time the report is right and has been emailed to the owner, it is after five. On a bad week, it is well after.

The report itself is useful. The owner reads it, and decisions are made from it. This piece is not about whether the report should exist. It is about the person who rebuilds it by hand every week, and what that ritual takes from them.

A task that is the same every week, done as if it were new

The striking thing about a manual weekly report is that almost nothing about it changes. The sources are the same. The steps are the same. The corrections are the same. Yet the work is done from scratch each time, because the steps live in one person's memory and hands rather than anywhere that can repeat them.

Spreadsheet and reporting tools have been built to deal with exactly this. Microsoft's documentation for Power Query, which is built into Excel and Power BI, describes the problem directly: shaping data by hand is one-off and not repeatable. Its answer is to record each transformation step, such as removing a column, fixing a data type or combining files, as a query that can be refreshed to get up-to-date data. Excel's own documentation describes refreshing data connections on demand, when a workbook opens, or every set number of minutes. Power BI's documentation describes scheduled refresh, with up to eight scheduled refreshes a day on a Pro licence.

Mentioning these tools is not a recommendation of one vendor. Other spreadsheet and reporting products offer similar features. The point is that the capability to make a report rebuild itself is widely available, and often already paid for, while the Friday ritual continues.

Why the ritual survives

Consider an illustrative case: a Polokwane building supplies business with about 30 staff. Five years ago the owner asked for a weekly summary. The administrator who set it up built it by hand, because that was the quickest way to get it out that week. The report grew a new tab every few months. Nobody ever had time to stop and rebuild it properly, because it had to go out every Friday.

The mechanism tends to have a few parts:

  • The first version was built in a hurry. Manual steps were faster than designing a repeatable process, and the report was needed that week.
  • The steps were never written down. Only the person who builds it knows which export to run, which rows to delete and which number is always wrong.
  • The work is invisible to the reader. The owner sees a finished report on Friday evening, not the three hours behind it.
  • Nobody owns improving it. Building the report is someone's job. Making it unnecessary to build is nobody's.

The decision side of this is covered in when the weekly report is already out of date, and the cost of waiting for figures in the hidden cost of waiting for management information. Here the focus is on the builder.

The human tax: time and attention

Time is the obvious cost: three hours a week is more than three working weeks a year, given to a task that is the same every time. Mellorca's guide to calculating the real cost of a repetitive task sets out how to put a figure on it.

Attention is the subtler cost. A fragile, manual process has to be carried in the head. The builder has to remember the sequence, the known errors and the workarounds. If they are ill on a Friday, the report either does not go out or someone else spends a stressful afternoon guessing. Some people in this position describe a low-level awareness of Friday that starts on Thursday. Mellorca cannot say how common that is, and it would be wrong to assume every builder feels it. Some enjoy the task, or value the visibility that comes with being the person who knows the numbers.

There is also a quieter effect on the end of the week. A report that has to be finished before anyone can leave pushes the last hours of Friday later, which connects to the question asked in Friday evening should start on Friday evening.

What could change

  1. Write the steps down first. Before automating anything, the builder lists every export, correction and check. This alone makes the report less dependent on one person.
  2. Remove steps nobody needs. Ask the owner which parts of the report are actually read. Tabs added years ago for a question that no longer matters can simply go.
  3. Make the shaping repeatable. Use a refreshable query to pull and clean each source, so that the corrections happen the same way every time without manual effort.
  4. Connect to the source where possible. Where systems allow a direct connection or scheduled export, the report can refresh without anyone exporting files.
  5. Change the builder's role. Once the report refreshes itself, the person who knows the numbers best can spend their time checking anomalies and explaining what changed, which is often the part the owner values most.

In the illustrative building supplies business, writing down the steps, dropping two unused tabs and turning the remaining sources into refreshable queries might turn three hours into twenty minutes of checking. The report would arrive earlier, and the administrator would leave on time.

Trade-offs

  • Upfront effort and skill. Building a refreshable report takes time the builder may not have, and some skill they may need to learn or borrow. Without that investment, the ritual continues.
  • Automated reports can fail quietly. Power BI's documentation notes that scheduled refresh can be deactivated after repeated failures, for example when credentials expire. A report that silently stops updating can be worse than one built by hand, so someone must check that it ran.
  • Manual assembly sometimes catches mistakes. A person pasting numbers notices when something looks odd. Automation should keep that checking step rather than remove it.
  • Identity and visibility. For some people, being the one who produces the report is part of how they are seen. Removing the task without recognising the knowledge behind it can feel like a loss rather than a gift.

Why it matters

The business case for an automated report is about speed and accuracy. The human case is about one person's Friday. When the report rebuilds itself, the builder gets back hours each week and no longer has to hold a fragile sequence in their head. Their knowledge of the numbers moves from assembling them to understanding them.

That is the dividend: a Friday afternoon that is no longer reserved for repetition, and attention freed for the part of the work that needs a person.

Sources and further reading

Related Mellorca reading

Article summaryEvery Friday afternoon one person exports, pastes, fixes and formats the same weekly report. What that ritual takes from their week and their attention, and what changes when the report builds itself.
Explore our solutionsStart a conversation