Beyond the map: designing an election explorer that reveals what actually mattered.
Most election coverage focuses on who won. Maps turn red and blue, seats change hands, commentators discuss swings and majorities. But after watching election after election, we found ourselves asking a different question.
Which votes actually changed the result?
Millions of Australians vote at every federal election, yet in practice, only a relatively small number of electorates determine who forms government. Many electorates are remarkably stable, reflecting long-term demographic, economic and cultural characteristics. Wealthier suburbs often lean conservative. Inner-city and university precincts frequently favour progressive parties. Rural areas develop their own distinctive political identities.
The election is decided not by every vote changing, but by the comparatively small number of places where voting behaviour changes enough to matter.
That became the central design principle for the Federal Election Explorer. Rather than simply displaying election results, we wanted to build an interface that helped explain where political change occurred, how significant it was, and whether it altered the national outcome.
Seeing Australia fairly.
Traditional electoral maps are geographically accurate. Unfortunately, they are often visually misleading.
Australia's federal electorates vary enormously in physical size. Urban electorates occupy only a few square kilometres, while remote electorates span hundreds of thousands of square kilometres. The electorate of Durack in Western Australia covers more than 1.4 million square kilometres—larger than many European countries combined.
A conventional map therefore gives enormous visual prominence to sparsely populated regions while compressing the dense metropolitan areas where most Australians actually live.
The problem is geographical, not electoral.
Every electorate elects exactly one member to the House of Representatives. From a visual design perspective it makes sense that every electorate should occupy approximately equal visual importance. This led to the cartogram.
A cartogram is a graphical representation of a geographic map that deliberately abandons the geographic size of territories in favour of electoral representation. In the Australian Federal Election, every electorate is shown at approximately the same size while preserving enough spatial relationships that Australia remains recognisable.
Creating a cartogram is part information visualisation and part graphic design. There is no perfect solution. Every electorate must be positioned carefully so the overall shape still feels like Australia while giving equal visual weight to every parliamentary seat.
When comparing the election results using the map of Australia versus looking at the cartogram, the cartogram feels simple and intuitive. Instead of seeing a large colour imbalance due to Australia's land mass and distributed population, you see a concise picture of how the results fell.
The importance of colour.
Colour presented another challenge. Most election maps simply colour electorates by the winning party. While useful, that tells only part of the story. Knowing who won is important, but it doesn't reveal how secure the victory was, how large the swing proved to be, or whether the electorate genuinely influenced the national result.
Rather than using colour to encode only party identity, we deliberately layered multiple electoral dimensions into the same visual language.
Using perceptually structured colour models, we independently control hue, chroma and lightness so each contributes different information without overwhelming the map.
In this case we used:
- Hue to identify the winning party or two-party-preferred result.
- Lightness to communicate electoral strength, allowing safe and marginal seats to be distinguished immediately.
- Chroma to reinforce analytical modes such as significance or swing.
This colour encoding approach allows the map to communicate both party identity and electoral consequence simultaneously.
Measuring significance.
One idea we found particularly interesting was distinguishing between swings that were merely statistical and swings that actually influenced the election.
Large swings occur in safe seats every election. They attract headlines, but often change nothing. Conversely, a relatively modest swing in a marginal electorate can determine who forms government. To capture this, the Explorer introduces significance.
Significance isn't simply measuring how much voting changed—it measures whether that change had electoral consequences. A seat experiencing a dramatic swing while remaining comfortably held is fundamentally different from a seat that changes hands after only a modest movement. Those are the electorates that reshape Parliament.
By highlighting electoral significance separately from swing and margin, the visualisation shifts attention towards the electorates in Australia that actually affected the result.
Watching the election unfold.
Election night is not a static event. Results arrive progressively throughout the evening, and the national picture gradually emerges from thousands of individual polling places. Replay reconstructs that experience, allowing users to watch the count unfold from the first declarations through to the final result. Rather than jumping directly to the outcome, viewers can revisit the uncertainty of election night and observe exactly when the national result became inevitable.
Looking inside each electorate.
Each electorate covers a significant geography, often spanning different neighbourhoods with very different political characteristics.
Polling booths offered an opportunity to explore that variation. Rather than treating polling places as isolated points on a map, we wanted to approximate the communities they serve.
The solution was to generate isochrones—equal travel-time catchments—around every polling location. These catchments are naturally imperfect. People don't always vote at their nearest booth, and community boundaries rarely follow drive-time contours. Nevertheless, they provide a reasonable spatial approximation of local voting communities and allow broader patterns to emerge. Instead of simply seeing one colour for an electorate, it becomes possible to observe the internal geography of political support and identify suburbs or regions whose voting behaviour differs from the electorate as a whole.
Parliament as the overall result.
The chamber view was intentionally added as the final view. Television broadcasters have long relied on parliamentary chamber diagrams because they immediately communicate the composition of Parliament and which party commands a majority.
The parliamentary chamber answers the simplest question: who has the numbers to govern?
The map answers a much richer one: how did Australia arrive there?
Together they connect electoral geography with the composition of Parliament.
Designing for exploration.
The Explorer contains considerably more information than can comfortably fit on a single screen. Every additional control competes for attention. Every legend occupies valuable space. Every interaction has to work equally well on desktop and mobile devices. Designing analytical tools is rarely about adding more information; it's often about deciding what not to show.
Throughout the project we aimed to create an interface that remained approachable despite the complexity of the underlying data. Party colours follow familiar conventions. Two-party-preferred results are always available. Parties are grouped according to Australia's broadly recognised political alignment, allowing relationships to be understood quickly without constant reference to legends.
The goal wasn't simply to build another election map; it was to create an environment where users could ask their own questions, follow their curiosity and gradually discover the patterns that emerge when elections are viewed through multiple complementary lenses.
Every vote matters equally. What differs is where those votes accumulate into political change. The purpose of the Explorer isn't to diminish individual votes, but to reveal the geography of those changes—where opinion shifted, where it remained remarkably stable, and ultimately which communities changed the direction of the country.