Most dashboards are built for stakeholders who want to see everything — not users who need to act on something.

Dashboards sit at the intersection of data, UX, and business intent — and all three usually want different things. Data teams want to show everything they can track. Product managers want to surface their metrics. Users want to know what to do next.

The result, in most products, is a page that shows a lot of numbers with no clear story. Users look at it, confirm that things are happening, and close it. Nothing gets acted on. This is the most common dashboard failure: a display of activity that doesn't enable a decision.

Before designing a single chart, the right question is: what decision does this dashboard support?

A dashboard that answers "how are things performing overall" is designed differently from one that answers "what should I do this morning" or "is something broken right now." Each implies a different data priority, a different level of detail, and a different interaction model.

Teams that skip this question end up designing a dashboard that tries to answer all three simultaneously — and therefore answers none of them well. Getting this question right is the UX Strategy work that makes the visual design decisions obvious.

More data is not more useful. Beyond a certain density, dashboards stop communicating and start overwhelming. Cognitive Load climbs, and users begin ignoring everything except the one or two numbers they already knew to check.

The discipline is curation. Every metric on a primary dashboard should be there because it drives a decision or surfaces a problem. Metrics that are "interesting" but not actionable belong in a drill-down view, not the primary surface. The most useful question to ask in a dashboard design review: if this number changed significantly, what would the user actually do differently?

A dashboard with no visual hierarchy forces users to re-prioritize everything every time they visit. Design the hierarchy around decision priority: the signal that needs the fastest response at the top or most prominent position, supporting context one level down, drill-down detail accessible but secondary.

Use Visual Hierarchy intentionally — not for decoration, but to guide where the eye goes and why. Color, size, and whitespace should reinforce information priority, not just create visual interest.

• Pie charts with more than four categories. Human perception can't accurately compare arc lengths past that point. Bar charts communicate the same data more clearly in almost every case.
• Numbers without baselines. A metric without context is meaningless. Users need to know whether 2,347 is up, down, or normal.
• Silent auto-refresh. Dashboards that change while a user is looking at them break focus. If real-time data matters, make updates visible — don't silently replace numbers mid-read.
• Tables where charts belong. Fifteen monthly revenue figures in a table require active mental processing. A bar chart communicates the trend in a second.
• Everything in red. Red signals urgency. When every metric below target turns red, users stop treating red as urgent and start ignoring it.