About
Every person who has ever been counted, in one picture.
We build tools that make population history legible: where people have lived, how fast places have grown or shrunk, and how confident we really are about numbers from before anyone was counting.
Why we exist
Most population numbers get told as a single figure. We think the story is in the shape: the flat centuries, the sudden climb, and the parts we're still only estimating.
How we work
Every number has a source
Each point on every chart is traceable to a named dataset and release year, not a black box.
Confidence is part of the data
Reconstructions, compiled history, estimates and projections are drawn differently, always.
History changes slowly, code should too
We rebuild from revised data on a schedule tied to how often the sources actually update.
Open by default
The data pipeline, sourcing notes and known gaps are documented, not hidden behind the chart.
Where the numbers come from
Three sources, stitched into one series
No single dataset covers 10,000 BCE to 2100, so we join three, in order, and mark exactly where one ends and the next begins.
Full sourcing notes, licenses and known seams are in the data documentation.
How to read the chart
Everything on the canvas has a meaning
The explorer packs a lot of signal into a small space. Here's what each visual layer tells you.
FAQ
Common questions
Where does the population data come from?
We join three open datasets: HYDE 3.3 for 10,000 BCE to 1799, Gapminder v7 for 1800–1949, and UN World Population Prospects 2024 for 1950–2100. Each source covers a distinct era and is clearly marked on the chart.
How accurate are the historical numbers?
Accuracy varies significantly by era. Modern figures (post-1950) are official UN estimates based on censuses. Pre-census numbers are scholarly reconstructions with wider uncertainty bands. The source band at the top of every chart shows which regime applies to each time period.
What is the difference between estimates and projections?
Estimates (up to 2024) are based on historical censuses, surveys, and vital registration systems. Projections (2025–2100) are UN median-variant forecasts derived from modelled fertility, mortality, and migration assumptions — not observations. They are drawn with dashed lines and a hatched background.
Why do some countries have data gaps?
Not every country has complete records for every year. Gaps may reflect missing census data, territorial changes, or periods of conflict that interrupted civil registration. The chart leaves those years empty rather than interpolating silently.
Can I download or use the data?
The underlying data comes from open sources: UN WPP 2024 (CC BY 3.0 IGO), HYDE 3.3, and Gapminder v7 — all under open licenses. You can export a CSV of the current chart view using the Export button in the explorer.
How often is the data updated?
We sync to the UN WPP on its release cycle (roughly every two years). HYDE and Gapminder revisions are less frequent. The footer of every page shows the current dataset revision date.
What does the cursor year control?
The cursor (the gold vertical line) marks the year the inspector panel reads from. Drag it or use the scrubber at the bottom to move through time. Values shown in the inspector always correspond to the cursor year.
Is there a mobile version?
Yes. The explorer adapts to small screens with a simplified canvas, a bottom sheet for filters, and a swipe-up inspector. Some advanced features (minimap, multi-country comparison) are desktop-only for now.