designJuly 22, 20252 min read

Complexity you can grow into

Powerful software doesn't have to show its power on the first screen. Let people meet it as they need it.

Open an export dialog and count the decisions it asks for. Format, quality, size and file name are fine. Then come colour profile, metadata, compression, upscaling and a naming pattern. Most days you only needed the first three.

The tool is right to have the rest, because someone needs them. The question is when they should appear.

Export4 of 12
FormatPNG
QualityHigh
Size2×
File namecover
BackgroundTransparent
Colour profilesRGB
MetadataStrip
CompressionBalanced
UpscalingOff
Naming pattern{name}@{size}
DitheringNone
Embed fontsOff
Four up front, the rest one click awayAn imagined export dialog, sorted by how often each option gets used.grid rows 0fr → 1fr, 240ms · rows fade in, 4px drop, 25ms apart · chevron 90° · ease (0.22, 1, 0.36, 1)

Show the few, keep the rest close

Jakob Nielsen described this in 2006 as progressive disclosure: show people a few of the most important options first, and offer the specialised ones on request. He connects it to learnability, efficiency of use and fewer errors, because newcomers aren't distracted by advanced settings and experienced people don't have to scan past them every time.

In the panel above, four options sit up front and the other eight are one click away. The extra rows open to their natural height without any measuring in JavaScript, because a grid row can animate from 0fr to 1fr:

.more {
  display: grid;
  grid-template-rows: 0fr;
  transition: grid-template-rows 240ms
    cubic-bezier(0.22, 1, 0.36, 1);
}
.more[data-open] {
  grid-template-rows: 1fr;
}
.more > div {
  overflow: hidden;
}

The rows inside fade in and drop 4px each, 25ms apart, so your eye travels down the new list instead of taking it in as one block.

Decide with evidence

The hard part is choosing what goes up front. Nielsen's advice is practical: task analysis and field studies for a new design, frequency-of-use statistics for an existing one. The usage behind this panel is invented, but the method holds. Sort by how often each option is used, then draw the line.

He also warns against stacking layers. Designs that go beyond two disclosure levels "typically have low usability," because people get lost moving between them. One "More options" is a doorway. Three nested ones make a maze.

Every visible choice costs a little

There's older research behind the instinct. Hick (1952) and Hyman (1953) found that choice reaction time grows with the information in a choice, roughly with the logarithm of the number of equally likely options. A settings panel isn't a reaction-time experiment, so I don't treat it as a formula for interfaces. I keep it as a reminder that each visible option asks for a little attention, even when you skip it.

Hiding things has a price too. Anything behind a disclosure is harder to discover, so the label on the doorway matters. "More export options" says what's inside. "Advanced" mostly says this part isn't for you.

The depth should be there on the day you need it and out of your way until then.