Poor data visualisation is immediately visible in an MBA assignment. A pie chart with 11 segments, a 3D bar chart that makes comparison impossible, or a chart titled "Figure 1" with no description — these signal that the student has produced graphics for the sake of it, not for analytical communication.
Principle 1: Chart Type Must Match Data Type and Question
There is no universally "best" chart. The choice depends entirely on what you're communicating:
- Bar chart (vertical or horizontal): Comparing quantities across categories (e.g., market share by competitor, sales by quarter). Most versatile and readable.
- Line chart: Showing change over time (e.g., revenue trend 2018–2024). Not appropriate for categorical comparison.
- Scatter plot: Showing the relationship between two continuous variables (e.g., advertising spend vs sales). Essential for regression visualisation.
- Stacked bar chart: Showing composition within totals over time or across groups. Use sparingly — can become unreadable with more than 4–5 components.
- Table: When exact values matter and the reader needs to compare specific numbers precisely. Not every set of data needs a chart.
Principle 2: Titles Should Convey the Finding, Not Just the Subject
A poor chart title: "Figure 3: Customer Satisfaction by Region" — this tells the reader what they're looking at, not what it means. A strong chart title: "Figure 3: Customer Satisfaction is Highest in the North East and Lowest in the Midlands (2024 Survey, n=320)" — this immediately communicates the analytical point. The reader should understand the key insight from the title alone.
Principle 3: Every Chart Needs a Source and Sample Size
In academic business reports, every chart must have a clear data source (e.g., "Source: Company Annual Report, 2023" or "Source: Author's survey data, n=145"). Missing sources are a referencing error. In quantitative research, include the sample size on charts derived from your own data.
Principle 4: Less Is More
Remove every visual element that does not carry analytical information: 3D effects, heavy gridlines, unnecessary borders, gradient fills, decorative shadows. These increase visual complexity without adding meaning. Excel's default chart styling is generally too "heavy" for academic reports — reduce to minimal styling.
Principle 5: Describe Charts in the Body Text
Never insert a chart and move on. Always write at least two sentences in the body text describing what the chart shows and what it means for your analysis. "As shown in Figure 3, customer satisfaction scores vary significantly by region, with the North East averaging 4.2/5.0 compared with the Midlands average of 3.1/5.0 — a difference that warrants regional strategy differentiation."