Short documentary · narration-led · portrait 9:16
How to make a documentary explainer video
A short documentary explainer works by visiting places in sequence and giving each one the same three pieces of information: where it is, one hard number, and why it matters. That repetition is what makes a chain of locations feel like an argument rather than a list. baz ships this as the Internet cables documentary skill — aerial flyovers between N locations, each stamped with a three-tier editorial pill system keyed to the voiceover, closing on a brand sign-off card.
The shortcut: run a skill instead of starting from scratch
A skill is a context file that teaches an AI agent one specific video style — its structure, pacing, typography, motion grammar and sound design. Hand an agent the skill plus your product, and it rebuilds that video for you. The steps below are what the Internet cables documentary skill already encodes, so you can follow them by hand or let the skill carry them.
baz is a library of video skills: context files that teach an AI agent a specific motion-graphics style, so it can rebuild that video for your product, with your brand, from your inputs. Run a skill inside your own coding agent — Claude Code, Codex — or directly in the baz editor.
What you need before you start
Required
- A voiceover script — this format is narration-led and requires one
- The location list, in the order you want them visited
Makes it materially better
- A quantified fact for every location — the format has a slot for it and an empty slot shows
- The tension claim for each: why this place is fragile, contested, or surprising
- Your channel identity for the sign-off card
How to make a documentary explainer video: step by step
Find the through-line before the locations
The locations are evidence for a claim, not the subject. "Most of the internet passes through a handful of places you have never heard of" is the video; the seven places are how you prove it. Without the claim you have a travelogue.
Give every location a hard number
The three-tier pill system has a slot for a quantified fact, and it is the tier that does the persuading. A location with a vague fact attached is the one the viewer scrolls past — find the number or drop the location.
Write the script first, then time everything to it
Every pill timing is keyed to a voiceover cue point and scene durations are set from the VO segments. Writing visuals first and recording narration later guarantees a mismatch you will have to fix by hand.
Order the locations to escalate
Each stop should raise the stakes on the last. If your most alarming fact comes third, the remaining four feel like an anticlimax. Save the strongest for the end.
Fly between locations rather than cutting
The aerial move between stops communicates the geographic relationship — how far apart, how concentrated, how exposed. Hard cuts throw away the one thing that makes a map-based documentary better than a list.
Verify every fact, then verify the borders
Geographic content attracts corrections faster than almost any other format. Check each claim against a real source, and be careful with any depiction of contested territory — a wrong border becomes the top comment and buries your argument.
What the finished video actually contains
Approximate timings for a ~61 seconds for seven locations cut. Treat them as a shape to hit, not a spec — the exact durations shift with your script and your music.
| Time | Beat | What is on screen |
|---|---|---|
| 0:00–0:05 | The claim | The through-line stated over the first aerial, before any location is named. |
| Per location | Three-tier stamp | Flyover arrives; pills land in sequence — place name, quantified fact, tension claim — each on its VO cue. |
| Between stops | Aerial transit | The camera travels, showing how the locations relate spatially. |
| Final stop | The strongest case | The location that proves the claim most sharply. |
| Close | Brand sign-off card | Channel identity and the follow prompt. |
Mistakes that cost you the view
Locations without a through-line
Do this instead: Lead with the claim. A sequence of interesting places with nothing connecting them is a travelogue, and it will not hold a portrait feed.
An empty fact tier
Do this instead: Every stop gets a number. The three-tier system makes a missing tier obvious, and the fact tier is the one carrying the persuasion.
Timing the visuals before recording the narration
Do this instead: Script and record first. Pill timings key to VO cue points — retrofitting the voice to finished visuals means re-timing every stamp.
Hard cuts between locations
Do this instead: Fly. The transit is what conveys distance and concentration, which is usually the actual point of the video.
Unverified geography
Do this instead: Check every border, name and figure. This audience corrects errors publicly and immediately.
Run it from your own agent
Skills are designed to run inside a terminal agent like Claude Code or Codex, against a project you control. The same skill produces the same style whether it runs in your agent or in the baz editor.
npm install -g bazaar.it@latest
# Install this skill into your agent (Claude Code, Codex or Cursor)
baz skill add geographic-chokepoint-explainer
baz project create --name "Internet cables documentary for my product"
baz prompt "make a documentary explainer video" --stream-json
baz review --summary --json
baz export start --wait --format mp4 --jsonOperations draw from a prepaid balance. See the CLI reference for the full command set.
Common questions
How many locations should I use?
Five to eight. The source example runs seven in about 61 seconds. Fewer and the pattern does not establish; more and each stop gets too little time to land its number.
Does it work for non-geographic subjects?
The three-tier stamp — name, number, tension — transfers to any sequence of things: companies, products, events. You lose the flyover, so you will need another transition to carry the relationship between items.
Do I need video editing experience?
No. The skill carries the editing decisions — shot order, pacing, type treatment, transitions, sound design. You supply the raw material and the facts about your product; the agent assembles the cut. You review it and ask for changes in plain language.
Can I run this in my own coding agent instead of the baz editor?
Yes. Skills are designed to be run from a terminal agent like Claude Code, Codex or Cursor. Install the CLI with `npm install -g bazaar.it@latest`, then `baz skill add <slug>` to pull the skill into your agent. The same skill produces the same style whether it runs in your agent or in the baz editor.
What does it cost?
baz bills against a prepaid dollar balance rather than a seat subscription — you top the balance up and each operation (generation, voiceover, render, export) draws a listed amount from it. There is no per-seat fee to make a video, and the price of each operation is published.
Can I change the video after it renders?
Yes, and this is the point of doing it as code rather than as a flat render. Every scene is a real component you can edit — ask the agent to re-time a beat, swap a screenshot, rewrite a line, or change the accent colour, and only that part rebuilds. You are not re-rendering from scratch each time.
Will every video from this skill look the same?
The grammar is the same — that is what makes it a skill and not a template. The content is not: your brand colours, typography, screenshots, copy and data all come from your inputs. Reusing one skill across a series is a feature, because episode three lands in the same visual language as episode one instead of being rebuilt from scratch.
Related guides
Pick a skill. Get the video.
Run the Internet cables documentary skill in your own coding agent, or in the baz editor. Usage comes out of a balance you top up, with a listed price per operation — no seat subscription to make a video.