Make Table Headers Carry Their Units

As of 2026-10-23 · The AZET team. We operate azet.io and are writing from that perspective. This article is a creative exercise for fun and reference. Its examples are invented, and the suggested AI workflow is an idea to try with a tool you already use, not a statement that AZET provides this capability.

A fictional workshop table has three columns: Item, Size, and Left. One row reads “Ribbon, 30, 4.” Is the ribbon thirty centimeters long, thirty millimeters wide, or sold in thirty-piece packs? Does four mean four rolls, four meters, or four remaining projects?

A neat table can still conceal basic uncertainty. This exercise uses AI to identify what a header fails to say. It does not ask the model to guess missing units or silently repair the records.

Separate the object from its measures

Write down what each column is intended to represent. For the invented ribbon example, suppose the source note says the ribbon is thirty millimeters wide and four rolls remain. Those are two different measures with two different units.

The revised headers can say “Width (mm)” and “Rolls remaining.” Keep the source note alongside the change. Without it, replacing “Size” with “Width” would be an unsupported interpretation.

If another row records sheets rather than rolls, do not force the count into the same label. Add a packaging field or separate the tables according to the reader's task.

Give AI permission to leave gaps

Use a prompt that values an unresolved cell over a confident invention:

Review this fictional table and its source notes. Identify ambiguous headers, mixed units, and values whose meaning cannot be determined. Propose clearer headers only where the notes support them. Mark unresolved cells for an editor. Do not convert units, infer packaging, or replace missing values with zero.

That last distinction matters even in a tiny exercise. Zero remaining rolls is a positive statement about the inventory. A blank cell may mean nobody checked. The table should preserve the difference instead of using typography to erase it.

Check the whole column, not one row

A header chosen from the first row may misdescribe later rows. Read vertically after making a change. If the “Width (mm)” column contains “small,” the table still mixes a physical measure with a category label.

Decide whether the category belongs in a separate column. Do not invent a numeric width for “small.” If the source cannot resolve it, keep a clear missing-value marker and a short explanation outside the table.

Likewise, inspect totals. Adding counts of rolls and sheets produces a number, but that number may not answer a useful question. A total should match the meaning of the column being totaled.

Ask a reader to explain one row

Give someone the revised table without the working notes. Ask them to describe the ribbon row in a sentence. Their explanation should preserve both the measure and its unit.

If they say “thirty rolls,” the revised design still leaves room for a mistaken association. Consider column order, spacing, or more explicit wording. Do not assume that a longer header alone solves the problem.

Save the table with its unit decisions and unresolved cells. When new rows arrive, use those decisions as an entry rule: a value needs the correct measure, unit, and source before it joins the column.

FAQ

Should every cell repeat the unit? Choose a consistent arrangement that readers can interpret. A unit in the header can work when the whole column shares it; mixed units need clearer treatment.

Can AI fill missing measurements? Not from an ambiguous label alone. The exercise requires source notes or a new measurement by the person maintaining the table.

Is a dash the same as zero? Only if you explicitly define it that way, which may confuse readers. Use distinct, explained markers for missing, inapplicable, and measured zero values.

A note from the makers

Our product reference is the maker introduction dated October 3, 2026. That document describes azet.io as a portal centered on the azet AI assistant and records an early-access email waitlist. It also says pricing was unpublished at that time. Those are dated statements from that introduction; this exercise does not establish a later launch, a feature release, or a pricing change.

If this kind of careful, everyday work interests you, visit azet.io to join the early-access waitlist. Read the information shown there before registering. We are sharing a way to think through a small task, with the final judgment left to the person doing it.