Data Visualisation: A Handbook for Data Driven Design: Summary & Key Insights

Every effective visualization begins with clarity—not of data, but of intent. The first question I always ask is, 'Why am I visualizing this data?' Without purpose, design becomes decoration. Purpose defines every subsequent choice: the choice of chart, the level of detail, the degree of interactivity.

Purpose is always shaped by context. Context includes where the visualization will appear (a printed report, a dashboard, a public website), who will view it, what they already know, and what they need to understand. A chart designed for executives may emphasize summary and clarity, while the same data for an analyst might be most valuable when it reveals nuance and invites exploration.

And then there is audience. Data visualization is not made for data—it is made for people. Different audiences interpret visuals differently, depending on their literacy in reading charts, their familiarity with the subject matter, and even cultural conventions of color or reading direction. Recognizing audience diversity means designing with empathy: choosing encodings that align with how users perceive and think, minimizing unnecessary complexity, and offering cues that guide interpretation.

When I work with clients, I frequently use what I call the 'why–who–how' reflection. Why are we visualizing? Who needs it? How will it be used? This framing pushes design away from mere aesthetics into intentional communication. A visual that works perfectly in one setting can fail entirely in another when context shifts. Staying attuned to purpose, context, and audience ensures that data visualizations remain adaptive, communicative, and humane.

No amount of design expertise can compensate for poor data understanding. In the visualization process, preparing the data is often the most time-consuming step. Before a single chart is drawn, we must truly 'meet' the data—understand how it was collected, what it measures, what it omits, and how reliable it is.

Data literacy begins with curiosity. In my own work, I always ask: what questions does the data attempt to answer? Which assumptions are embedded within it? Cleaning and transforming data—removing errors, standardizing formats, handling missing values—is not glamorous, but it is the foundation of trustworthy visualization.

Equally important is understanding data types and relationships. Whether values are categorical, ordinal, or continuous directly determines which visual encodings make sense. For example, color hue is best for categorical comparisons, while position and length work better for quantitative data. Misalignment between data type and encoding is one of the most common causes of misleading visual communication.

This preparatory stage is not about perfection but about clarity. I encourage designers to keep a notebook of data observations: patterns that surprise, limitations that matter, stories that might emerge. By developing a close relationship with your data, you’re not merely a designer—you’re a translator of reality.