Power bi tips data visualisation best practice
Download as PPTX, PDF0 likes178 views
The document outlines best practices for data visualization in dashboards, emphasizing the importance of planning, understanding the audience, and choosing appropriate chart types based on data. It provides guidance on using space effectively and refining designs through user feedback, while also presenting examples of well and poorly constructed dashboards. Key principles include maintaining design consistency, ensuring clarity, and avoiding common mistakes such as clutter and lack of context.
Power bi tips data visualisation best practice
- 3. Presentation contents: • Have a PLAN • Know your audience • What type of dashboard? • Data Types • Use Appropriate Charts • Make best use of space • Refine the design • Examples of Good and Bad
- 4. Have a PLAN Follow a standard process as much as possible when introducing new dashboards. Try to keep consistency with design, layout and flow. Purpose • Audience • Main Question Layout • Main Dashboard • Supporting Reports • Drill Downs Analysis • Understand the data • Get user feedback New Release • Training • Documentation • Support
- 5. Know Your Audience Understand the needs of the end user. Design the dashboard at the relevant level for that audience. e.g. a CEO will be likely want a high level overview, whereas a HR manager will tend to be focussed on people, skills and training.
- 6. Types of Dashboards Strategic High Level, Low interactivity, Long strategic timeframes. Analytical Highly interactive, Self service, Ability to explore data with context e.g. Historical or Budgets. Operational Process Monitoring, Routines, Immediate timeframes.
- 7. Types of Data Categorical / Nominal Named items, with no relative values. Examples might include: • Countries • People • Departments Quantitative Items that have a measured value. Examples: • Sales £ • Prices • Time Ordinal Named items that have a value based order. Examples: • Premium, Standard, Budget • High, Medium, Low
- 8. Use Appropriate Charts Recommended Charts Bullet graph – horizontal type gauge, more space efficient. Bar graphs – great for comparatives, space efficient, used for comparatives.
- 9. Use Appropriate Charts Recommended Charts Stacked Bar Chart – useful to combine headline view with a breakdown into categories. Combination bar and line – useful to show individuals values and the trend using the line at the same time. 0 2 4 6 8 10 12 14 2018 2017 2016 2015 UK Europe USA 47% 48% 49% 50% 51% 52% 0.0 1.0 2.0 3.0 4.0 5.0 2015 2016 2017 2018 Sales Cost of Sales Gross Margin %
- 10. Use Appropriate Charts Recommended Charts Sparklines – very compact space efficient object, to give a general trend feel, to give some context, used to decide whether to look in more detail. Box Plots – useful to present distribution of data around a median, with upper and lower values, e.g. defect rates.
- 11. Use Appropriate Charts Recommended Charts Treemaps – space efficient way of breaking down a hierarchy by values, to prompt further exploration in more detail. Pie Charts – useful to show breakdown of the whole, but not as space efficient as bar charts or stacked bar charts. There is a belief that Pie Charts are not the best presentation medium, But, they can be a very useful tool to ‘filter’ data and they are widely understood.
- 12. Use Appropriate Charts Recommended Charts Tables If used correctly, a table can often be clearer to present the data than a chart. . City Actual Budget London 5.1 5 Birmingham 3.2 3.3 Manchester 2.4 2 Sheffield 1.6 1.5 Maps Can be very useful to measure relative geographic data, which may be difficult to see any other way, the example below shows sales cannibalisation in Sheffield from a new store opening in nearby Rotherham.
- 13. Make the best use of Space Space is a significant limiting factor. Important to try to keep to a familiar standard. Similar to reading left to right, top to bottom. Very Important •Main Message •Key Measures •Don’t use for logos Secondary Importance •Use for sub reporting / analysis Secondary Importance •Use for sub reporting / analysis Least Important •Use for least relevant parts •Good for further sub analysis or drill through •Good location for a logo Most important area
- 14. Refine Building the perfect dashboard is not something that can be done without an iterative approach. With the right PLAN methodology you can get off to a good start, with the basic structure. The key element is to learn what your audience needs and continue to refine, if the end user doesn’t immediately understand what the dashboard is telling them then it probably needs to change.
- 15. Example of a badly laid out dashboard This dashboard does present a lot of information, Sales, Budgets, Labour costs. It is just so difficult to take the information on board. There is no flow, its too busy to read quickly.
- 16. Example of a badly laid out dashboard This dashboard does present a lot of information, Sales, Budgets, Labour costs. It is just so difficult to take the information on board. There is no flow, its too busy to read quickly. Too many decimal places Scroll Bars can be frustrating Prime space, wasted with a logo Good, but needs to show values or use tooltips Not space efficient, generally substance over form Difficult to make relevant comparisons Can’t see the categories, too many colours Lacking context, and difficult to compare values to anything Looks good, but of limited usefulness
- 17. Mistakes to avoid • Fragmenting Data • Lack of Context • Over Elaboration • Too many decimal places • Using too much space for little data value • Too much clutter • Unnecessary graphics
- 18. Example of a better dashboard This dashboard is much easier to follow. Clear sections encourage right to left reading. Lack of clutter allows the data to be more quickly understood.
- 19. Best Practice - Recap • Have a PLAN • Know your audience • What type of dashboard? • Data Types • Use Appropriate Charts • Make the best use of space • Refine with user feedback
- 20. Hope you found this useful
- 21. For more Tips, Tricks and Timesavers, visit our website Tips and Timesavers | Select Distinct Limited Credit: simon.harrison@selectdistinct.co.uk