They say a picture is worth a thousand words - well, in data visualisation, a good chart might save you a thousand meetings. Data visualisation is where art meets science, turning raw numbers into a story that even your busiest stakeholder can understand. In this article, we’ll explore the golden rules of data visualisation - guidelines that ensure your visuals are not just eye-catching but also insightful and effective.
Let’s turn data chaos into visual clarity, one golden rule at a time!
1) Audience and Purpose:
Tailor visuals to the knowledge level and goals of the audience.
Be clear on whether the report is for exploration (discovery of insights), explanation (telling a story), or monitoring (keeping track of KPIs).
Use a Narrative Structure - the report should have a beginning, middle, and end.
2) Simplicity and Clarity:
Less is more - essential data only, avoid too many details in a single view. Include just the visuals that add value; avoid excessive use of colours, shapes, or irrelevant graphics. Ensure all visuals have descriptive titles, labels, and annotations (tooltips for details) where necessary.
Avoid visual/page scrolls and use non-standard report sizes.
3) Choose the Right Chart Type:
Bar Charts - for comparisons.
Line Charts - for trends over time.
Pie and Donut Charts - only for parts of a whole (limit to 3-5 categories).
Scatter Plots - for relationships between two variables.
Maps - for geographic data.
Cards – for single metric display (KPI) and/or Target versus Actual.
Avoid Common Pitfalls:
Using different colours for the same categories across visuals can confuse users; maintain colour consistency throughout.
Mixing units (e.g., thousands vs. millions) or using inconsistent number formats across visuals can confuse users; ensure consistency in units and formatting.
Avoid 3D charts, as they often distort data.
Stacked bar charts can be difficult to read when there are many segments or when small differences need comparison; avoid them if precise comparisons are required.
Changing axis scales inconsistently or omitting zero can create misleading interpretations; ensure axis scales are logical and consistent.
Be cautious with pie charts and complex visuals like radar charts; these can be difficult for users to interpret.
Using too many pie charts or including too many slices within a single pie makes it hard to compare values accurately.
4) Colour Use and Accessibility:
Use colour to differentiate categories, highlight key data, or show progression. Limit colour use for emphasis only.
Use colour theory - red and green for negatives and positives, respectively.
Accessibility - ensure colorblind-friendly palettes. Adding labels and legends is always advised.
Avoid Color Overuse - stick to 3-5 colours to maintain a clean look.
Make sure to follow the organisation's colour and brand guidelines.
5) Organise Layout and Visual Flow
Flow from Left to Right, Top to Bottom - organise visuals in a way that naturally leads the eye, with the most important information at the top.
The Z-pattern is also very common – people are naturally drawn to the flow that starts at the top left corner, moves horizontally to the right, then diagonally to the bottom left, and finally moves horizontally to the bottom right, creating a Z-shape.
Group-related visuals.
Maintain consistent spacing and alignment.
6) Interactive Elements
Filters, Slicers and Drill-Throughs should be used for exploring data in more detail.
Slicers are a great way of allowing users to navigate data, but they come at a performance cost, so thoroughly evaluate their needs and remove the unnecessary.
Place the page filters on the right side, as most people naturally use their computer mouse with their right hand, making right-aligned filters more intuitive and accessible, and also ensuring a smoother interaction.
Setting up filters that don’t interact correctly across visuals can lead to conflicting data views and unexpected results or data confusion. Make sure filters are synchronised where needed.
7) Data Accuracy and Validation
Ensure that the data source is reliable and that the data is current. Regularly check that your visuals and metrics reflect accurate calculations and consistent page formatting.
Display the last data refresh date prominently for reports with time-sensitive information.
8) Performance Optimisation
Limit visuals on each page - cramming too many visuals onto a single page affects readability and slows down performance. The rule of thumb is limiting the number of visuals to 10-12, with 6–8 visuals per page ideal for clarity, focus, and performance. If more visuals are necessary, split them into multiple pages or use tabs to organise the content logically.
Avoid importing excess columns or tables that won’t be used in the report—only load necessary data. For larger datasets, consider aggregating data at a higher level before bringing it into Power BI.
Optimise DAX queries - use efficient DAX measures, avoid unnecessary complex calculations, and use variables where possible.
9) Documentation
Maintain a data dictionary or glossary for any terms and acronyms used in the report.
Provide brief descriptions of custom metrics, especially complex ones.
And there you have it, the golden rules to transform your data into visuals that inform, engage, and inspire. The true magic of data visualisation lies in its ability to bridge the gap between raw numbers and meaningful narratives.
As you embark on your next chart or dashboard, ask yourself: Are you telling the story your data deserves? Or better yet, how will you use these golden rules to level up your visualisations? Let’s turn data into insights and insights into impact—one golden rule at a time.
______
by Margarida Pereira
@ Passio Consulting
Comments