- ✓Effective data visualisation makes complex data accessible to audiences with varying levels of technical expertise.
- ✓The choice of chart type should match the type of data and the story you want to tell; bar charts suit comparisons, line charts show trends, scatter plots reveal relationships.
- ✓Dashboards aggregate multiple visualisations to provide a real-time overview of key metrics, enabling faster and better-informed decisions.
- ✓Tools such as Power BI, Tableau, and Python's matplotlib and seaborn libraries each have different strengths and are suited to different use cases.
- ✓Data storytelling combines accurate visualisation with clear narrative to guide the audience from raw data to insight to action.
Listen to the full episode inside the course. Enrol to access all 80 episodes, plus assignments, tutor support and Student Finance funding.
Start learning →Alex: Today we're covering data visualisation and storytelling. Sam, why does how you present data matter as much as the data itself?
Sam: Because humans are visual creatures. We process visual information much faster and more intuitively than tables of numbers. A well-designed chart can communicate a pattern, trend, or anomaly in an instant that would take minutes to extract from a spreadsheet. But a poorly designed chart can mislead, confuse, or even conceal the very insights it's supposed to reveal.
Alex: What are the principles of good data visualisation?
Sam: The first principle is to choose the right chart type for what you're showing. Bar charts are for comparing quantities across categories. Line charts are for showing change over time. Scatter plots show the relationship between two continuous variables. Histograms show the distribution of a single variable. Pie charts are useful only when you have very few categories and you're showing parts of a whole. Never use a 3D chart; they distort perception and add no information.
Alex: What are some common mistakes?
Sam: Truncating the axis of a bar chart to exaggerate differences. Using colour to encode information that colourblind people can't distinguish. Overloading a single chart with too much information. Using misleading scales. Not labelling axes clearly. And perhaps most commonly: presenting a chart without any narrative to explain what the reader should take from it. A chart should always come with a clear message.
Alex: What tools are available for building visualisations?
Sam: For a computing student, Python's matplotlib and seaborn libraries provide programmatic control over every element of a chart and are suitable for analytical and report contexts. Plotly adds interactivity. For business dashboards, Power BI and Tableau are the dominant tools in most organisations: they allow non-technical users to build and publish dashboards connected to live data sources. Each tool has a different learning curve and different appropriate use cases.
Alex: What is data storytelling and how does it differ from just presenting charts?
Sam: Data storytelling weaves data, visuals, and narrative together into a coherent story that takes the audience from a question to an insight to an action. The narrative is the crucial element that charts alone lack. A good data story has a clear beginning that establishes the context and the question; a middle that presents the evidence, building understanding progressively; and an end that states the conclusion clearly and identifies what should happen next. The audience should leave knowing exactly what was found and what it means for them.
Alex: Any advice on making visualisations accessible to non-technical audiences?
Sam: Use plain language in all labels and titles. State the insight in the chart title rather than just describing what the chart shows: 'Sales increased 23% in Q4' is more useful than 'Quarterly Sales Data'. Minimise the amount of text on the chart itself. Use colour purposefully and consistently. And always provide context: what is this number compared to, what was it last year, what does success look like?
Alex: Brilliant. Thanks Sam. We start Unit 11 next, on Systems Analysis and Design.