Dashboards

A useful dashboard answers a small set of recurring questions at a glance, then gives people enough context to explore further. Start with the audience and decision—not the available charts.

Define the dashboard's job

Before building, write down:

• Who will use it
• What decision it should support
• Which measures define success or risk
• How often the data needs to refresh

For example: “Help regional sales leaders spot weekly revenue changes and decide where to investigate.”

Ask Dawn for a first cut

Add the relevant dataset or analysis to a conversation and describe the audience, decision, and required measures.

Create a dashboard outline for regional sales leaders. Lead with weekly revenue,
year-on-year growth, and conversion rate. Include enough detail to identify which
regions and products explain a change, but keep the first view suitable for a five-minute review.

Treat the first version as a proposal. Check that the measures are defined correctly and that each visual has a clear purpose.

Design for scanning

• Put the most important decision signals first.
• Use a small number of measures on the opening view.
• Give charts descriptive titles that state what the reader should notice.
• Make comparisons explicit: versus target, prior period, forecast, or another segment.
• Use filters only when they unlock a real question.

Review before sharing

Validate totals against the source data, test the most important filters, and ask a representative user to complete a real task with the dashboard. If a chart does not help them answer a question, simplify it or remove it.

Keep it useful over time

When data refreshes regularly, connect the dashboard to a reusable pipeline and review the metric definitions whenever the business changes. Use a report when the audience needs written interpretation alongside the numbers.