How we make our 60-second 3D explainers

Each Big Picture 60 explainer packs one big story into about a minute: seven scenes, a handful of numbers and a clear question. Here is how we build them, and the rules every figure has to pass.

By Big Picture 60 · Published 19 October 2026 · 4 min read

1. One question per scene

Every scene follows the same pattern: a header states the claim, a 3D visual shows the evidence, and a short callout gives the takeaway. Each line of narration carries at most one or two statistics, so a viewer can follow it in a single listen.

2. Every figure is sourced and dated

• Each scene shows its source line on screen, and every video ends with a sources card listing all citations.
• Estimates are marked "est."; contested figures are labelled "disputed" and attributed to whoever made the claim.
• Large numbers are written the readable way — "$1.2 trillion", not "$1,200,000,000,000".
• Charts join only sourced data points; we never draw a smooth line through data we don't have.
• Current events carry an "as of" date, casualty counts are labelled incomplete, and schematic visuals (such as strike routes) are marked illustrative.
• We never fake official documents. When we show a policy record in 3D, it is labelled a reconstruction, paraphrased from the cited source, with no seals or stamps.

3. Real maps, not drawings

Maps come from Natural Earth's 1:50m country boundaries (public domain). A script projects each region once, keeps enclaves and islands that matter, simplifies outlines to the detail visible on screen and draws internal borders. Countries shown side by side for a comparison that is not about area are labelled "not to scale".

4. Narration and sound

The narration is read by a synthetic neural text-to-speech voice, with a quiet ambient music bed mixed to YouTube's loudness standard. The script the voice reads is the on-screen narration, so what you hear and what you read never drift apart, and each line is timed to the scene it describes.

5. Built in 3D, rendered as video

Each explainer is an interactive Three.js web page you can orbit and pause. The same page is rendered frame by frame into a 16:9 video and a 9:16 YouTube Short, so the website, the video and the Short always show the same scenes.

6. How we use AI

We use AI tools to help research, write code and draft scripts. Every published figure is tied to a named source and date that you can check yourself, and we correct errors in place, noting the date of the change.