Conditional Formatting on Widgets

Conditional formatting is the highest-return five minutes you will ever spend on a dashboard. It takes a chart of neutral blue bars and turns it into something a reader can interpret in one glance: red is a problem, green is fine. No legend, no explanation, no training.

It lives in the widget's properties panel, on a measure, and it works on bar charts, values tables, and number widgets alike.

Where to find it

Select the widget you want to colour, open the right-hand properties panel, and look for the Conditional Formatting section. Switch it on and you get one required choice: which measure drives the colour. A chart usually has just one, but if your query returns several, be deliberate — you might colour the row by margin while the bar length shows revenue.

Bins and thresholds

A bin is just a value range with a colour attached. The default gives you three, which is almost always what you want, because most business rules are three-state: below target, near target, above target.

Say you're colouring sum of amount on an opportunity chart. You might set:

The thresholds are yours to choose, and this is where the thinking actually happens. Don't pick round numbers because they're round. Pick the number that means something to the business — the quota, the renewal floor, the margin below which a deal isn't worth doing. A threshold that matches a real decision boundary makes the colour actionable; an arbitrary one just makes the chart colourful.

Rename the labels

Out of the box, each bin is named something forgettable like Bin 1. Overwrite them. Type Low, Medium, and High — or better, the words your team actually uses: At Risk, On Track, Exceeding.

This matters more than it looks, because those labels surface in the widget's legend and in tooltips. A legend reading "Low / Medium / High" is self-documenting. One reading "Bin 1 / Bin 2 / Bin 3" makes the reader ask you what it means, which is exactly the question good design removes.

Bins versus gradient

Discrete bins are the default, but charts also offer a gradient option. Instead of three hard buckets, the colour shifts continuously across the value range — pale at the low end, saturated at the high end.

Which to use comes down to whether your rule has an edge:

• Bins when there's a real cut-off. Quota attainment, SLA breach, pass/fail. The reader needs to know which side of the line each value sits on, and a hard colour change communicates that instantly.
• Gradient when you're showing relative magnitude with no meaningful boundary. Revenue by region, activity counts, heat-style comparisons. A gradient shows the shape of the distribution without pretending there's a threshold that doesn't exist.

A good gut check: if you can name the threshold in a sentence, use bins. If you find yourself inventing a number to justify a cut-off, use a gradient.

It works on number widgets too

Conditional formatting isn't only for charts. Apply it to a number widget and the big figure itself changes colour — a single KPI that turns red the moment it drops under target. That's a very efficient use of dashboard real estate: one number, no legend, no chart, and the state is unmistakable from across a room.

We'll build exactly that kind of formatted number widget when we assemble the KPI row later in this section.

Now that your charts carry meaning through colour, the next problem is letting users slice them without hunting through a filter panel. Next up: date toggles and faceting.