Short excerpt from a YouTube talk arguing dashboards must provide contextual comparators (e.g., prior period, budget) to make raw numbers meaningful, and that color should be used sparingly in visualizations to reduce cognitive load and draw attention only to important items.
Chapters
If you just showing a number on the dashboard we made a million dollars last month. Is that good or bad, right? It depends on what it was last year, depends on what the budget number was. How are you going to look at a number and evaluate whether that performance is good or bad? And that's where you need the context. So how are you going to be able to evaluate it? What's the context we need to show here? Those are the kind of insights that you need to get up front. If you bake those into your report it will be much, much better. And then the next thing is thinking about data visualization, right? Reducing cognitive load. Not like color just by default creating a chart and you have all these rainbow color charts because they don't help people to be able to understand the data and what it's telling them. So actually using color really sparingly and using it to draw attention to things that you want to draw attention to that are important.