Skip to content
UX Atlas

Design problem

Dense dashboards

Twenty charts, no argument. The dashboard reports numbers without telling anyone what to do.

How you know you have it

  • Nobody can say which metric the dashboard exists to move.
  • Every widget is the same size, so every metric looks equally important.
  • Charts use a rainbow palette and a legend three scrolls away.
  • People export to a spreadsheet to do the actual comparison.

Do these first

  1. 1Write the three questions this dashboard answers, and delete widgets that answer none of them.
  2. 2Give the primary metric more size, weight and space than the rest, deliberately.
  3. 3Add a comparison to every number: previous period, target, or benchmark.
  4. 4Label series directly instead of sending the eye to a legend.

Why it happens

Built from available data rather than from questions

A dashboard assembled from what the warehouse exposes answers nothing in particular.

No density decision

Comfortable spacing for a monitoring wall and compact spacing for an analyst are different products. Picking neither produces both badly.

Comparison left to the reader

Numbers without a baseline, a target or a previous period cannot be interpreted, only read aloud.

What to read, and why

Dashboards fail on hierarchy, density and chart choice more often than on data, so these entries cover the three together.

Why it behaves this way

5

The mechanism underneath the symptom. Read these when the fix needs justifying to somebody else.

  • Visibility of System Status

    The interface should always tell the user what is going on, through timely and appropriate feedback.

    HeuristicExpert heuristic

  • Cognitive Load

    Working memory is a narrow channel, and interfaces spend it on three kinds of work, only one of which is useful.

    PsychologyEstablished research

  • Recognition Over Recall

    Recognising something on screen is far easier than remembering it, so keep the information visible.

    HeuristicExpert heuristic

  • Miller's Law

    Working memory holds a small number of chunks, so the design lever is what counts as a chunk, not the raw item count.

    LawEstablished research

  • Anchoring Bias

    The first number a person sees shapes every judgement that follows, even when it is obviously arbitrary.

    Cognitive biasEstablished research

The craft decisions

7

The visual and structural levers that produce or relieve this problem.

  • Density

    How much information occupies a given area, which should follow the user's expertise and task rather than a house style.

    UI foundationCraft convention

  • Visual Hierarchy

    The order in which elements are noticed, controlled by size, weight, colour, position and space.

    UI foundationCraft convention

  • Scale

    Relative size is the fastest signal of importance and the easiest one to overuse.

    UI foundationCraft convention

  • Grids

    A shared structure that makes alignment automatic and layout decisions repeatable.

    UI foundationCraft convention

  • Whitespace

    Empty space is a design element that groups, emphasises and sets pace, not leftover area.

    UI foundationCraft convention

  • Data Visualization Color

    Charts need three kinds of palette, and the choice between them is determined by the data, not by taste.

    UI foundationCraft convention

  • Typographic Hierarchy

    Using size, weight, colour and space to make the structure of text visible before it is read.

    UI foundationCraft convention

Patterns that address it

5

Recognised solutions, with the conditions under which each earns its place.

  • Filtering

    Narrowing a large set by attributes, which is what makes large catalogues navigable at all.

    PatternCraft convention

  • Sorting

    Reordering a set, where the default order is the decision that matters most.

    PatternCraft convention

  • Empty States

    The screen before there is anything to show, which is a teaching opportunity rather than a gap.

    PatternCraft convention

  • Loading States

    What to show while waiting, chosen by how long the wait is and whether its length is known.

    PatternCraft convention

  • Comparison

    Putting options side by side so differences can be seen rather than remembered.

    PatternCraft convention

Components involved

5

States, behaviour and the accessibility contract you will need to specify.

  • Chart

    A visual encoding of data, which is only useful if the encoding matches the question being asked.

    ComponentEstablished research

  • Data Table

    The right tool for comparing many records across many attributes, and a large amount of behaviour to specify.

    ComponentCraft convention

  • Tabs

    Parallel views in one space, which works only when users never need to compare across them.

    ComponentCraft convention

  • Date Picker

    One of the hardest components to get right, and one where typing is often faster than picking.

    ComponentCraft convention

  • Tooltip

    A short description on hover or focus, which must never be the only place information lives.

    ComponentCraft convention

Language

2

Where the words carry more of the fix than the layout does.

  • Microcopy

    The small pieces of text that do most of the work: labels, hints, empty states, errors and confirmations.

    UX writingCraft convention

  • Content Design

    Deciding what content a user needs and in what order, which is a design activity rather than a writing one.

    UX writingCraft convention

Accessibility requirements

3

The requirements this problem raises, mapped to the decisions that produce them.

  • Use of Color

    Colour may reinforce meaning but must never be the only thing carrying it.

    AccessibilityCraft convention

  • Keyboard Navigation

    Everything must be operable with a keyboard alone, which is the single most load-bearing accessibility requirement.

    AccessibilityCraft convention

  • Contrast

    Difference is what makes anything perceivable, and it is measurable for text against its background.

    UI foundationCraft convention

How to find out

2

Methods suited to confirming the cause before you commit to a fix.

  • User Interviews

    Structured conversations about what people actually did, which is far more reliable than what they say they will do.

    Research methodEstablished research

  • Product Analytics

    Measuring what people do at scale, which tells you where problems are and never why they exist.

    Research methodCraft convention

Check your work

1

Interactive review passes, with progress saved in your browser.

  • Dashboard UX Checklist

    Whether a dashboard answers a question or just displays numbers.

    ChecklistCraft convention

Decisions other teams made

1

Worked cases, including the options that were rejected.

Where this usually shows up

The same material, organised around the surface you are building rather than the symptom you are fixing.