Why this matters
Great dashboards do two things: make the most important information impossible to miss and make the next step obvious. As a Data Analyst, you will:
- Present exec scorecards where one KPI drives decisions.
- Build monitoring views so operators catch issues fast.
- Create diagnostic pages that guide users from overview to detail.
Layout and hierarchy turn the same charts from “confusing collage” into “clear story.”
Progress saving note
You can read, practice, and take the quick test here for everyone. If you’re logged in, your progress and test results will be saved automatically.
Concept explained simply
Layout is how you arrange elements on the page. Hierarchy is how you signal what matters most using position, size, contrast, and grouping.
- Reading patterns: In left-to-right languages, eyes start at the top-left and scan in an F-pattern for dense content and a Z-pattern for lighter pages.
- Zones: Give the top-left the primary KPI; use the first row for key trends; place diagnostics and details below.
- Grouping: Proximity, alignment, and shared titles make related items feel connected.
- Whitespace: Space is a design tool; it creates separation and focus.
- Progressive disclosure: Overview → causes → details. Don’t show everything at once.
Mental model
Think of your dashboard like a newspaper front page:
- Headline (top-left): the primary KPI or status.
- Lead story (first row): 2–3 big trend/variance charts that explain the headline.
- Sections (below): diagnostic breakouts, tables, and filters.
Users should get the headline in 5 seconds, the story in 30 seconds, and the detail in 3 minutes.
Key principles you can apply today
- Use a grid (e.g., 12 columns). Align cards to the grid to avoid messy gaps.
- Prioritize by position: highest priority = top-left; second = first row center; third = first row right.
- Prioritize by size: bigger cards feel more important—but only for 1–3 items.
- Typography scale: Title > KPI number > axis labels. Keep it consistent.
- Color restraint: Gray for normal, a single accent for highlights/alerts.
- Filters placement: Put global filters in a left sidebar or top bar; avoid scattering them across cards.
- One idea per card: If it needs a paragraph to explain, split it into multiple cards.
Common layout patterns
Executive scorecard
- Row 1: Big KPI cards (Revenue, Margin, Growth) with sparklines and target deltas.
- Row 2: Category/segment variance bars and win/lose drivers.
- Row 3: Exceptions table (only outliers) and footnotes/definitions.
Monitoring / operations
- Left: Vertical filter/controls panel.
- Row 1: Status tiles with alert colors; overall throughput chart.
- Row 2: By-queue or by-region small multiples.
- Row 3: Live incidents table sorted by severity.
Diagnostic explorer
- Row 1: Overview trend + decomposition (e.g., stacked bars).
- Row 2: Dimension switcher (tabs) and the focused breakdown.
- Row 3: Drill table with conditional formatting.
Worked examples
Example 1: Sales performance (monthly)
- Goal: Execs decide if targets are on track.
- Hierarchy: Top-left: Total Revenue vs Target; First row: YoY trend and Margin %; Second row: Segment variance bar and Top 5 drivers; Third row: Exceptions table.
- Why it works: Decision is visible in 5 seconds; drivers in 30 seconds.
Example 2: Support operations (real-time)
- Goal: Team leads act on SLAs and backlog.
- Hierarchy: Top-left: SLA status tile; First row: Live queue size and first response time; Second row: Small multiples by channel; Third row: Incidents table.
- Why it works: Alerts first, then trend, then where to act.
Example 3: Marketing diagnostics
- Goal: Analysts find channels causing CAC spikes.
- Hierarchy: Top-left: CAC vs target; First row: CAC trend and Spend vs Leads; Second row: Channel ROAS small multiples; Third row: Drill table with channel x campaign.
- Why it works: Clear overview with fast path to channel-level detail.
Step-by-step workflow
- Define decisions and audience: What decision happens on this page?
- Inventory metrics: List KPIs (primary, secondary), diagnostics, filters.
- Prioritize: Rank by importance and frequency.
- Wireframe: Sketch zones (headline, first row, diagnostics, table).
- 5-second test: Ask a peer what the page is about after 5 seconds.
- Build: Use consistent card sizes and typography.
- Refine: Remove one thing per iteration that doesn’t help the main decision.
Quick checklist
- Primary KPI at top-left, large, with target delta.
- 1–3 charts in first row tell the story.
- Filters are grouped (left or top), not scattered.
- One accent color for alerts/highlights only.
- Whitespace used intentionally between sections.
Exercises
Do these on paper or in your BI tool. Keep notes; you’ll compare with solutions.
Exercise 1 — Prioritize a sales dashboard
Audience: CFO, monthly review. Inputs: Revenue, Margin %, CAC, Pipeline coverage, Top drivers, Segment breakdown, Exceptions table, Filters (Region, Product).
- Task: Arrange cards into three zones: Headline, Story, Diagnostics. Write a one-paragraph description of your layout and why your top-left item deserves that spot.
Exercise 2 — Fix a cluttered ops dashboard
Audience: Support lead, hourly. Problem: 10 charts crammed; no clear status; filters appear on multiple cards.
- Task: Redesign using the Monitoring pattern. Define an alert-first first row, place filters, limit colors, and decide what moves to the table. Write the rules you applied.
Compare with solutions
See the Exercises section below for detailed solutions with reasoning.
Common mistakes and how to self-check
- Everything looks important: Too many big charts. Fix by promoting only 1–3 items; reduce others.
- Scattered filters: Users don’t know what’s applied. Fix by centralizing global filters.
- Color overload: Multiple bright hues reduce emphasis. Fix by using grayscale baseline with a single accent.
- Inconsistent card sizes: Creates visual noise. Fix by using a grid and a small set of card sizes.
- Headline without context: KPI with no target or trend. Fix by adding delta vs target and a sparkline.
Self-check in 60 seconds
- Can a new user state the dashboard’s purpose after 5 seconds?
- Is the primary KPI top-left and largest?
- Are there at most 3 items in the first row?
- Are filters in one place and clearly labeled?
- Is there enough whitespace to see sections?
Practical projects
- Project 1: Redesign a company KPI page into a 3-row scorecard with a clear headline and exceptions table.
- Project 2: Build a monitoring dashboard for a fake warehouse (throughput, delays, incidents) with an alert-first first row.
- Project 3: Create a diagnostic explorer for a marketing dataset with channel tabs and a drill table.
Who this is for
- Data Analysts, BI Developers, and anyone building dashboards that must drive decisions.
Prerequisites
- Basic chart literacy (bar/line/table)
- Familiarity with a BI tool (e.g., grid layout, cards, filters)
Learning path
- Before: KPI definition and data modeling basics
- Now: Dashboard layout and hierarchy
- Next: Interaction design (filters, drill-down) and alerting
Next steps
- Finish the exercises and take the quick test.
- Apply the redesign to an existing dashboard and request 5-second feedback from two users.
- Iterate once, removing at least one non-essential element.
Mini challenge
You have 20 minutes to sketch a weekly executive scorecard. Include: 1 headline KPI with target delta, 2 trend charts, 1 drivers chart, and an exceptions table. Ensure the headline is top-left and the filters are grouped. Optional: add a brief subtitle that states the decision the page supports.