Dashboard Design, UX and Data Storytelling

Introduction

A dashboard is not merely a collection of charts. It is an information interface designed to help a particular audience understand a situation, identify exceptions, and decide what to do next. In Power BI, effective dashboard design combines data analysis, visual communication, user experience (UX), accessibility, and organizational context.

A technically correct report can still fail if users cannot find the important information, misunderstand a metric, or are overwhelmed by unnecessary visuals. Conversely, a visually attractive dashboard may be ineffective if it lacks trustworthy definitions, appropriate comparisons, or a clear relationship to business decisions.

This chapter explains how to design dashboards for executive, managerial, and operational audiences. It also examines layout, navigation, accessibility, responsive and mobile design, annotations, narrative, and context. The emphasis is practical: the goal is to create Power BI experiences that communicate insight clearly and support responsible action.

Learning Objectives

By the end of this chapter, you should be able to:

1. Distinguish the information needs of executive, managerial, and operational users.
2. Apply principles of visual hierarchy, layout, UX, and dashboard composition.
3. Select suitable charts and interactions for different analytical tasks.
4. Design navigation structures for Power BI reports.
5. Apply accessibility and mobile design principles.
6. Communicate insights using annotations, narrative, comparisons, and context.
7. Evaluate a dashboard for clarity, usability, accuracy, and ethical risk.

1. Dashboards as Decision-Support Interfaces

A dashboard connects data to decisions. It normally performs four related functions:

• Monitoring: showing the current state of important measures.
• Exploration: allowing users to investigate patterns and segments.
• Diagnosis: helping users understand why a result occurred.
• Action: directing attention toward a decision, intervention, or follow-up.

These functions should not be confused. A monitoring dashboard may require a small number of prominent indicators, whereas an exploratory report may require filters, drill-through pages, and detailed tables. A single report can contain all four functions, but each page should have a clearly defined purpose.

A useful design question is: What decision should this page help the user make? If the answer is unclear, the page may be a data display rather than a decision-support tool.

Dashboard, Report, and Scorecard

The terms are often used interchangeably, but they describe different emphases:

Power BI reports can support all three forms. The distinction is therefore conceptual rather than a strict software limitation.

2. Designing for the Audience

The same data should not be presented identically to every user. Audience design begins with the user’s responsibilities, decisions, time available, analytical skill, and tolerance for detail.

Executive Dashboards

Executives generally need a concise view of organizational health, strategic progress, risk, and material exceptions. An executive page should typically answer:

• Are we meeting strategic objectives?
• What has changed since the previous period?
• Where is performance materially above or below expectation?
• What risks or opportunities require attention?

Executive dashboards should emphasize a limited set of high-value KPIs, trends, targets, and exceptions. They should avoid dense operational tables and excessive slicers on the landing page. Definitions and supporting detail should still be available through tooltips, linked pages, or drill-through.

Managerial Dashboards

Managers usually need to compare teams, products, regions, channels, or processes. They need enough detail to diagnose performance and allocate resources. A managerial dashboard may include:

• Actual versus target performance.
• Period-over-period trends.
• Variance by business unit.
• Contribution analysis.
• Filters for responsible teams, products, or locations.

Managers benefit from controlled interactivity. Every filter should have a business purpose; a page with many unstructured slicers increases cognitive load and can make results difficult to reproduce.

Operational Dashboards

Operational users need timely, actionable information about current work. Examples include open orders, service-level breaches, stock shortages, production defects, or unresolved support cases. Operational dashboards may require:

• More granular records.
• Refresh-time information.
• Queues and exception lists.
• Clear ownership fields.
• Drill-through to individual transactions.

For operations, a visually simple table with conditional formatting may be more useful than a sophisticated chart. The design should make the next action obvious.

3. Visual Hierarchy and Layout

Visual hierarchy determines the order in which users notice and interpret information. It is created through position, size, contrast, whitespace, alignment, colour, typography, and repetition.

A practical dashboard layout often follows this sequence:

1. Orientation: title, reporting period, filters, and definitions.
2. Summary: key performance indicators and status.
3. Pattern: trends, comparisons, and distributions.
4. Explanation: drivers, segments, or exceptions.
5. Action: detail, ownership, commentary, or next steps.

This structure moves from “What is happening?” to “Why?” and finally to “What should happen next?”

The Grid and Alignment

A grid helps users scan a page because related objects appear organized and predictable. Visuals should normally share edges, use consistent spacing, and align with other elements. Uneven spacing can make a report feel unreliable even when the calculations are correct.

Use whitespace deliberately. Empty space separates groups, establishes importance, and prevents visual crowding. Filling every available area with charts does not increase analytical value.

One Page, One Primary Purpose

A page should have a dominant question. For example:

• “Are sales on track this quarter?”
• “Which regions are causing the margin decline?”
• “Which service cases require immediate action?”

A page attempting to answer unrelated questions often becomes a dashboard “wall,” where no visual receives sufficient attention.

KPI Cards

KPI cards are useful for prominent measures such as revenue, margin, customer retention, or open cases. A good KPI card includes:

• A clear label.
• The current value.
• A comparison baseline, such as target or prior period.
• A directional indicator where appropriate.
• The period or scope of the value.

A value without a comparison is often ambiguous. For example, “£2.4 million” cannot be judged without knowing whether it is above target, seasonal, or declining.

4. Choosing Effective Visuals

Chart selection should follow the analytical task rather than personal preference.

Bar charts are usually preferable to pie charts when precise category comparison matters. Pie and donut charts can communicate simple part-to-whole relationships, but they become difficult to read with many categories or similar-sized segments.

Three-dimensional effects, decorative images, excessive gradients, and unnecessary gauges often reduce clarity. Visual decoration should never compete with the data.

Data-Ink and Signal-to-Noise

A useful design principle is to maximise the visual emphasis placed on meaningful data and minimise non-informative elements. Borders, gridlines, background shapes, and labels should be retained only when they improve interpretation. This does not mean removing all structure; it means ensuring that structure serves a communicative purpose.

5. Context, Comparisons, and Interpretation

Numbers become meaningful through context. A dashboard should help users interpret a measure by providing appropriate comparisons, definitions, and time frames.

Common comparison types include:

• Actual versus budget or target.
• Current period versus previous period.
• Current period versus the same period last year.
• One segment versus another.
• Performance relative to a benchmark.

The comparison must be valid. Comparing a partial month with a complete month can create a misleading conclusion. Similarly, year-over-year comparisons may be distorted by changes in product mix, pricing, calendar structure, or data coverage.

Variance

A basic absolute variance is:

Here, Actual is the observed value and Reference may be a target, budget, forecast, or comparison period. The units are the same as the measure.

A percentage variance is:

This calculation assumes that the reference value is not zero. If the reference is zero or close to zero, the percentage can be undefined or misleading; an absolute variance or an alternative business rule should then be used. For measures where lower is better, such as defects or response time, the interpretation of a positive variance must be explicitly labelled rather than assumed to be favourable.

Definitions and Metadata

A dashboard should make important definitions discoverable. Users may interpret “sales,” “customer,” “profit,” or “on-time delivery” differently. Include a glossary, tooltip, information button, or linked methodology page explaining:

• Measure definition.
• Inclusion and exclusion rules.
• Data refresh time.
• Currency and units.
• Time zone and reporting period.
• Target source and ownership.

6. Navigation and Interaction in Power BI

Navigation should help users form a mental model of the report. A common structure is:

1. Overview or executive summary.
2. Performance analysis.
3. Driver or segment analysis.
4. Operational detail.
5. Definitions and methodology.

Power BI features that support this structure include buttons, page navigators, bookmarks, drill-through, tooltips, slicers, and cross-filtering.

Navigation Principles

• Keep the primary navigation in a consistent location.
• Use meaningful labels such as “Regional Performance,” not “Page 3.”
• Show the selected page or state clearly.
• Provide a visible route back to the overview.
• Avoid hiding essential navigation behind unexplained icons.
• Ensure keyboard users can move through interactive elements logically.

Drill-Down and Drill-Through

Drill-down reveals lower levels within a hierarchy, such as year to quarter to month. Drill-through takes the user to a separate page filtered to a selected context, such as a customer or product. These interactions should be predictable and signposted. A user should understand what changed and how to return.

Tooltips

Tooltips can provide definitions, secondary measures, or small supporting visuals without overcrowding the main page. They should supplement, not conceal, information necessary for the primary decision.

7. Accessibility and Inclusive Design

Accessibility is a quality requirement, not an optional enhancement. A dashboard may be used by people with visual, motor, cognitive, or learning differences, as well as users working in poor lighting, on small screens, or under time pressure.

Important practices include:

• Do not use colour as the only method of communication.
• Use sufficient contrast between text and backgrounds.
• Provide descriptive titles and alternative text where supported.
• Use meaningful tab order.
• Label axes and units clearly.
• Avoid very small fonts.
• Use patterns, icons, labels, or position alongside colour.
• Test with keyboard navigation and screen-reader-compatible structures where applicable.
• Avoid red–green combinations as the sole distinction because colour-vision deficiencies are common.

A status indicator might use colour, a symbol, and text: “Below target,” “On target,” or “Above target.” This is more robust than relying on colour alone.

Accessibility also includes cognitive clarity. Consistent terminology, restrained interaction, predictable navigation, and explanatory notes reduce the mental effort required to use the report.

8. Responsive and Mobile Design

A desktop dashboard cannot simply be compressed onto a phone. Mobile users have less screen space, may use touch rather than a mouse, and often need rapid access to a small number of decisions.

A mobile layout should prioritise:

• The most important KPI cards.
• A small number of high-value trends.
• Clear vertical reading order.
• Large touch targets.
• Short labels and readable text.
• Minimal horizontal scrolling.
• Essential filters only.

Power BI mobile layouts should be tested on realistic devices rather than assumed to work from the desktop canvas. Check whether titles wrap, labels overlap, slicers remain usable, and interactions are understandable with touch.

Responsive design is also relevant to embedded reports and changing browser windows. A design that works only at one fixed size is fragile. Use consistent spacing, simple structures, and deliberate visual priorities.

9. Data Storytelling

Data storytelling combines evidence, explanation, and audience relevance. It is not the practice of manipulating data to create a dramatic narrative. A responsible story makes the evidence easier to understand while preserving uncertainty and limitations.

A practical storytelling sequence is:

1. Situation: What context does the audience need?
2. Change: What has happened?
3. Significance: Why does it matter?
4. Drivers: What factors appear associated with the change?
5. Action: What decision or investigation follows?

For example, a report might show that customer cancellations increased during a quarter. The narrative should identify the comparison period, quantify the change, distinguish observed facts from hypotheses, and indicate what analysis is required next.

Annotations

Annotations draw attention to significant points, such as:

• A target breach.
• A policy change.
• A supply interruption.
• A product launch.
• An unusual data gap.

An annotation should be brief, specific, and evidence-based. It should explain relevance rather than merely repeat the value displayed in a chart.

Narrative Text

Narrative text can be placed in subtitles, text boxes, dynamic titles, or commentary panels. Strong narrative answers questions such as:

• What is the key message?
• What evidence supports it?
• What should the user examine next?

Avoid presenting an interpretation as a fact when the dashboard only shows correlation. “Sales fell after the price change” describes sequence; “the price change caused the fall” is a causal claim requiring stronger evidence.

10. Applied Case: Northstar Retail Performance Report

The following is an illustrative case.

Background

Northstar Retail operates 85 stores and an online channel. Its existing Power BI report contains revenue, transactions, product categories, and regional results. Users complain that the report is slow to interpret and that different departments use different definitions of “sales performance.”

Problem

The executive team wants a rapid view of revenue, margin, and strategic targets. Regional managers need to identify underperforming stores and understand category drivers. Store managers need an actionable list of stock and service issues. The existing report places all these requirements on one crowded page.

Analysis

A requirements workshop identifies three audiences and three decision levels:

The team also discovers that online sales are refreshed daily, while some store data is refreshed hourly. This refresh difference must be disclosed.

Approach and Solution

The redesigned report uses five pages:

1. Executive Overview: six carefully selected KPIs, target comparisons, a revenue trend, and an exception panel.
2. Regional Performance: a sorted bar chart, regional trend, category contribution, and controlled slicers.
3. Store Diagnosis: drill-through from region to store, with actual-versus-target analysis.
4. Operational Actions: an exception table showing stock shortages, overdue service cases, owner, and due date.
5. Definitions and Data Quality: metric definitions, refresh information, and known limitations.

The executive page uses neutral colours for ordinary values and a high-contrast accent for exceptions. The report does not use red and green alone; labels such as “Below target” and “Above target” are included. An annotation explains that online data is updated daily and store transactions hourly.

Outcome

Executives can identify the largest performance risks without navigating operational detail. Managers can move from a regional variance to affected stores through drill-through. Store managers receive an actionable queue rather than a collection of summary charts. The redesign also reduces disagreement because definitions and refresh times are visible.

Learning Points

The case demonstrates that dashboard improvement is primarily a problem of audience alignment and information architecture. Adding more visuals would not have solved the original problem. Separating decision levels, clarifying definitions, and creating purposeful navigation produced greater value.

11. Governance, Ethics, and Quality Assurance

Dashboard designers influence what users notice and how they interpret organizational performance. This creates ethical responsibilities.

Accuracy and Transparency

Numbers should be traceable to governed data sources. Report owners should document transformations, refresh schedules, exclusions, and known quality problems. A dashboard should not imply real-time accuracy if the underlying data is delayed.

Fairness and Privacy

Operational dashboards may expose employee, customer, or patient information. Apply row-level security and data minimisation. Display only the level of personal detail necessary for the legitimate business purpose. Aggregation may be preferable where individual identification creates unnecessary risk.

Avoiding Misleading Design

Misleading practices include truncated axes without clear justification, inconsistent scales, selective time periods, hidden filters, ambiguous denominators, and colour coding that exaggerates small differences. The issue is not whether a design is technically permissible; it is whether a reasonable user could reach a materially false interpretation.

Testing Checklist

Before publication, test the dashboard for:

• Correct calculations and filter behaviour.
• Clear measure definitions.
• Appropriate comparisons and time periods.
• Readability at normal viewing distance.
• Accessibility and keyboard order.
• Mobile layout and touch interaction.
• Navigation and drill-through return paths.
• Performance under realistic data volumes.
• Security and user permissions.
• Consistency between displayed narrative and underlying evidence.

12. Contemporary Developments

Modern BI platforms increasingly support natural-language questions, automated narratives, anomaly detection, and AI-assisted visual creation. These features can accelerate exploration but do not replace governance or human judgement. AI-generated explanations may omit business context, confuse correlation with causation, or rely on poorly defined measures.

Use AI assistance as a starting point for investigation. Validate every generated insight against the model, definitions, filters, and source data. Clearly distinguish automated observations from verified managerial conclusions.

The continuing movement toward embedded analytics also changes dashboard design. Reports may appear inside customer portals, operational systems, or collaboration tools. In these settings, the report must fit the surrounding workflow, respect permissions, load efficiently, and provide a clear transition from insight to action.

Chapter Summary

Effective dashboard design begins with the decision and audience, not with the chart catalogue. Executive dashboards emphasise strategic status and exceptions; managerial dashboards support comparison and diagnosis; operational dashboards support immediate action. Layout, alignment, whitespace, hierarchy, and purposeful visual selection make information easier to scan.

Good dashboards provide context through targets, time comparisons, definitions, units, and refresh information. Power BI navigation features such as buttons, drill-through, bookmarks, tooltips, and slicers should create a predictable information structure. Accessibility requires more than colour coding: contrast, labels, alternative text, tab order, readable typography, and cognitive clarity all matter. Mobile design requires deliberate prioritisation rather than simple shrinking.

Data storytelling combines situation, change, significance, drivers, and action. Annotations and narrative should clarify evidence without overstating causality. Finally, trustworthy dashboards require governance, privacy protection, quality assurance, and critical evaluation of AI-assisted features.

Key Terms

Dashboard: An interactive interface that presents selected information for monitoring, analysis, or decision-making.

Data storytelling: The communication of evidence through data, visual structure, explanation, and context.

Executive dashboard: A concise dashboard focused on strategic performance, risk, and exceptions.

Managerial dashboard: A dashboard supporting comparison, diagnosis, resource allocation, and performance management.

Operational dashboard: A dashboard supporting current work, exception handling, and immediate action.

KPI: A key performance indicator used to assess progress toward an important objective.

Variance: The difference between an actual value and a reference value such as a target or budget.

Benchmark: A standard or comparison point used to interpret performance.

Visual hierarchy: The ordering of attention created by size, position, contrast, spacing, and typography.

Cognitive load: The mental effort required to understand and operate an interface.

Drill-down: Movement from a higher level of a hierarchy to a lower level within a visual.

Drill-through: Navigation to another report page filtered to the selected context.

Tooltip: Supplementary information displayed when a user points to a visual element.

Accessibility: The design of information and interaction so that people with diverse abilities can use it.

Responsive design: Design that adapts effectively to different screen sizes and devices.

Annotation: A concise explanatory note attached to a data point, event, or visual.

Data quality: The extent to which data is accurate, complete, timely, consistent, and fit for purpose.

Row-level security: A control that restricts the rows visible to a user according to defined permissions.

Review Questions

Short Answer Questions

1. What is the difference between a dashboard and an analytical report?
2. Why should a dashboard provide comparisons rather than isolated values?
3. Name three features in Power BI that can support report navigation.
4. Why is colour alone insufficient for communicating status?
5. What is the purpose of an annotation?

Long Answer Questions

1. Explain how the information needs of executive, managerial, and operational users differ. Give suitable dashboard elements for each audience.
2. Discuss how visual hierarchy, grid alignment, whitespace, and page purpose affect dashboard usability.
3. Explain how context and definitions influence the interpretation of KPIs and variances.
4. Describe a practical approach to designing an accessible and mobile-friendly Power BI report.
5. Discuss the ethical risks associated with misleading visual design and unsupported data narratives.

Analytical / Application Questions

1. A sales dashboard shows current revenue of £480,000 against a target of £500,000. Calculate the absolute and percentage variance, and explain how the result should be presented to an executive audience.
2. Design a page structure for a customer-service dashboard used by both senior managers and service-team supervisors. Explain which information belongs on the overview page and which belongs on an operational page.
3. A report claims that a price increase caused a decline in sales because the decline occurred after the price change. Evaluate this conclusion and propose additional analysis needed before making a causal claim.

Multiple Choice Questions with Answers

Question 1

Which dashboard is most focused on immediate exception handling and current work queues?

A. Executive dashboard

B. Operational dashboard

C. Strategic scorecard

D. Methodology report

Correct Answer: B. Operational dashboard

Question 2

What is the main purpose of visual hierarchy in a dashboard?

A. To maximise the number of charts on a page

B. To ensure every visual has identical emphasis

C. To guide the order in which users notice and interpret information

D. To replace the need for measure definitions

Correct Answer: C. To guide the order in which users notice and interpret information

Question 3

Which Power BI feature takes a user to another page filtered to a selected context?

A. Drill-through

B. Data refresh

C. Bookmark formatting

D. Conditional formatting

Correct Answer: A. Drill-through

Question 4

Why should colour not be the only way to communicate whether a KPI is on target?

A. Colour always slows report performance

B. Some users may not distinguish colours reliably

C. Colour cannot be used in Power BI

D. Colour is visible only on mobile devices

Correct Answer: B. Some users may not distinguish colours reliably

Question 5

Which visual is generally most suitable for comparing many categories precisely?

A. Bar chart

B. Decorative image

C. Gauge with no target

D. Three-dimensional pie chart

Correct Answer: A. Bar chart

Question 6

A measure has an actual value of 90 and a reference value of 100. What is the percentage variance using ((Actual − Reference) ÷ Reference) × 100%?

A. −10%

B. 10%

C. 90%

D. 110%

Correct Answer: A. −10%

Question 7

Which statement best describes responsible data storytelling?

A. It exaggerates patterns to increase engagement

B. It presents evidence and context without overstating causality

C. It removes all uncertainty from the narrative

D. It replaces data definitions with persuasive language

Correct Answer: B. It presents evidence and context without overstating causality

Question 8

What should a mobile dashboard generally prioritise?

A. Every desktop visual at a smaller size

B. The most important KPIs and a small number of high-value interactions

C. Long tables requiring horizontal scrolling

D. Hidden filters and decorative graphics

Correct Answer: B. The most important KPIs and a small number of high-value interactions

Question 9

What is a key reason for documenting data refresh times?

A. To make all data appear real-time

B. To help users understand the timeliness and limits of the information

C. To eliminate the need for security controls

D. To increase the number of report pages

Correct Answer: B. To help users understand the timeliness and limits of the information

Question 10

Which practice creates a misleading dashboard?

A. Showing the reporting period

B. Providing a target definition

C. Using hidden filters that materially change results

D. Labelling units on an axis

Correct Answer: C. Using hidden filters that materially change results