I design tools for people who work with data, and I write about how they get made.

Last spring I spent three weeks on a research vessel off the Bohuslän coast, watching marine biologists use an early build of Tidepool. I went to learn about offline sync. I came back thinking about empty states.

The first thing I noticed was the tide table taped inside the wheelhouse. Every row was full. Even at slack water, when nothing much is happening, the table says so: a time, a height, a direction. Nobody on board ever looked at a cell and wondered whether the printer had failed.

Blank is not a value

Our app did the opposite. When a sensor dropped out, when a sample had not been logged yet, or when a reading was genuinely zero, the screen showed the same thing: nothing. Three very different situations, one empty rectangle. The biologists filled the gap with guesses, and the guesses were usually pessimistic. They assumed data was lost when it was only waiting for a signal.

An empty state is not the absence of content. It is content about absence, and it deserves the same care as everything else on the screen.

Naming the kinds of nothing

Back on land, we changed the data model so that every missing value carries a reason. A reading can be measured, pending, unavailable or not applicable, and each one gets its own treatment on screen.

{
  "station": "Smögen",
  "time": "10:31",
  "height_m": null,
  "status": "pending",
  "reason": "waiting_for_signal"
}

The interface change was small. Pending readings show a dashed outline and the time they were captured. Unavailable readings name the sensor and suggest a fallback. A real zero is printed as 0.00, in the same weight as every other number, because zero is a measurement and should look like one.

What I do differently now

I ask for the empty states in the first week of a project, not the last. For each screen I list every way it can be empty, write the sentence a person needs to read in that moment, and only then start drawing. It is slower at the beginning and much faster at the end.

The tide table got there first, of course. It has been answering the question “what is happening right now, even if it is nothing?” for well over a century. Most of my job is learning to be as honest as a sheet of paper taped to a wall.