What makes a KPI dashboard different
A KPI dashboard is not an analytical report. An analytical report is designed for exploration — a user arrives with a question, navigates through pages and filters, and leaves with an answer. A KPI dashboard is designed for monitoring — a user arrives to confirm that everything is on track, or to identify immediately if something is not.
This distinction drives every design decision. An analytical report can have multiple pages, many visuals per page, and complex filtering options. A KPI dashboard should ideally fit on one screen, require no interaction to deliver its primary message, and communicate its most important numbers within five seconds of opening.
The anatomy of a business KPI dashboard
A well-structured business KPI dashboard has four visual layers, arranged in order of importance from top to bottom. The top layer holds headline numbers — the three to five most critical metrics the business tracks, displayed as large card visuals that are visible from across the room. The second layer holds trend context — how those headline numbers compare to last year, expressed as gauge indicators or growth percentage cards. The third layer holds supporting detail — a live scroller, a summary table, or a map that adds context without demanding focus. The fourth layer, if present, holds drill-down access — links or buttons to the analytical reports where users can explore further.
In a travel platform business context, the headline numbers are: total bookings YTD, total business revenue YTD, total room nights YTD, supplier bookings all years, and booking agent count. These are the five numbers that senior leadership checks first, every time they open the dashboard.
A complete KPI dashboard published to Power BI Service — headline cards, gauges, and a map, all on one screen.
Card visuals for headline numbers
Card visuals display a single number prominently — no axes, no bars, just the number and a label. They are the correct visual for headline KPIs because they communicate one thing with maximum legibility. A number shown as a bar chart requires the reader to read the axis and estimate the value. A number shown as a card is read instantly.
The key design decisions for card visuals are size, font weight, and colour. Make the number large — at least three times the size of the label text. Use a heavy font weight so the number stands out against the background. Use colour deliberately: green for metrics that are above target or growing, red or amber for metrics that are behind, and a neutral colour for metrics where current-vs-target context is not available.
In Power BI, conditional formatting on card visuals allows the number colour to change automatically based on its value relative to a threshold. A booking growth card that turns green when YTD is ahead of LYTD and amber when it is behind removes the need for a reader to calculate the implication — the colour communicates it instantly.
Gauge indicators for target tracking
A gauge visual displays a value on a curved scale between a minimum and maximum, with the needle or fill showing where the current value sits relative to a target. It communicates not just the value but its position within a defined range — which is the question that matters for target tracking: are we on track, ahead, or behind?
In a travel platform KPI dashboard, gauge indicators show booking growth YTD versus LYTD and business growth YTD versus LYTD. The minimum is zero (no growth), the target is the planned growth rate, and the maximum is set above the target to give headroom. When the gauge needle sits above the target marker, the business is outperforming plan. When it sits below, it is underperforming.
Gauge visuals are most effective when the target value is meaningful and well-understood by the audience. A gauge without a clear target is just a semicircular bar chart — the shape implies a target exists even when it does not, which misleads readers. Only use gauges when you have a genuine, agreed target to show.
In Power BI, the gauge visual has three value fields: Value (the current metric), Minimum Value, and Maximum Value, plus a Target Value that appears as a marker on the scale. Set these using DAX measures rather than fixed numbers so the gauge updates automatically as data refreshes.
The live scroller ticker
A live scroller — sometimes called a ticker — is a horizontally scrolling strip of data that cycles through a list of values continuously. In Power BI, this is implemented using a custom visual or a text-based marquee element on a dashboard tile. It adds a sense of real-time activity to a dashboard without taking up significant screen space.
In a travel platform KPI dashboard, the live scroller shows the top hotels by YTD bookings, cycling through hotel names and their booking counts. It sits just below the headline card visuals, occupying a single narrow horizontal band. A reader who wants to know which hotels are performing best can read the scroller passively while the rest of the dashboard holds their primary attention.
The scroller is a supporting element, not a primary one. It should never contain the most critical information on the dashboard — critical information belongs in static card visuals where it is always visible. The scroller is for contextual detail that adds richness without demanding focus.
Map visuals for geographic distribution
A map visual plots data geographically, with bubble size or colour intensity representing a metric value at each location. In Power BI, the built-in map visual uses Bing Maps to geocode location fields — city names, country names, or coordinates — and places bubbles whose size reflects the measure value.
In a KPI dashboard for a travel business with branches across multiple cities and countries, the map visual shows all bookings and booking growth by branch, with bubble size representing YTD booking count and bubble colour representing growth versus last year. A reader can immediately see which geographic areas are the platform\'s strongest markets and which are growing fastest — information that would take multiple table rows to communicate and seconds to extract from a map.
The map visual requires clean, consistent location data. City names that are spelled inconsistently (Mumbai vs Bombay, UAE vs United Arab Emirates) will geocode to different locations or fail to geocode at all. Standardising location field values in Power Query before loading to the model is essential for reliable map rendering.
Layout and visual hierarchy
The layout of a KPI dashboard should follow the reader\'s natural eye movement pattern — top left to right, then down. The most important information belongs in the top-left and top-centre of the canvas. Secondary information belongs below and to the right. The bottom of the canvas is for supporting context that readers access intentionally rather than absorbing automatically.
Visual hierarchy is reinforced through size, weight, and whitespace. Large card visuals with bold numbers draw the eye first. Smaller gauge indicators below them provide context. The map or scroller at the bottom provides depth. Sufficient whitespace between visual groups prevents the dashboard from feeling cluttered and helps each element breathe.
Avoid the temptation to use every pixel. A dashboard that is 80% filled with visuals is harder to read than one that is 60% filled, because the reader\'s eye has no clear entry point and no visual resting places. Whitespace is not wasted space — it is the structure that makes the content legible.
Keeping it to one screen
The one-screen constraint is the hardest discipline in KPI dashboard design. Every stakeholder has a metric they consider critical. Every team wants their numbers on the headline dashboard. The PM or BI practitioner who owns the dashboard must make and defend the prioritisation decision: what are the five to seven numbers that, if known, tell you whether the business is healthy?
Everything else belongs on analytical report pages that are linked from the dashboard. The KPI dashboard is the front door — it answers "are we okay?" The analytical reports answer "why?" and "what specifically?" Conflating the two produces a dashboard that answers neither question clearly.
Interactive KPI dashboard preview
The interactive layout below demonstrates the visual hierarchy of a business KPI dashboard — headline cards at the top, gauge indicators in the middle, supporting map and scroller below. Click any element to see the design rationale for that visual choice.
In the next post in this series, we look at choosing the right visual for each analytical question — the decision framework that prevents the most common dashboard design mistakes.
`