How to Make an Explainer Video for YouTube

Most explainer videos on YouTube do not explain anything. They summarise a topic. You finish watching with a general sense of the subject but no clearer answer to the specific question the title implied. The reason is almost always structural: the creator started with a topic instead of a question.

A real explainer answers one specific question, and everything in it serves that answer.

Start with one question, not a topic

An explainer about "the Federal Reserve" is a topic video. An explainer that answers "why did the Federal Reserve keep rates high for 18 months after inflation started falling" has a thesis to build and resolve.

Pick the question before you write a single line of script. The question determines what research to do, which data to include, and where the video ends. Without it, a 15-minute video drifts through related facts with no clear resolution, and viewers drop off before you reach one.

The best explainer titles on YouTube are often questions or specific claims: "Why the UK is poorer than every US state," "How one man caused the 1997 Asian financial crisis," "Why Argentina keeps collapsing." Each one makes a promise the video has to keep.

Script structure

Open with the question or the problem, not the answer. Give the viewer a reason to need the explanation before you start giving it. A 30-second setup that establishes why the question matters pulls viewers through the next 12 minutes more reliably than front-loading the conclusion.

Use concrete numbers and named examples throughout. "Inflation was high" is inert. "US inflation peaked at 9.1% in June 2022, the highest reading since 1981" gives the viewer something to hold onto.

Write for spoken delivery. Short sentences. Active verbs. Read every paragraph aloud before you finalise it. If you run out of breath mid-sentence, the sentence is too long for a script.

Structure the middle of the video as a sequence of sub-questions. Each section answers one piece of the main question and sets up the next. The viewer should always know where they are in the argument and what is still unresolved.

Visuals: what actually explains things

Talking head footage builds personality and trust. It does not carry information by itself. A face on screen while a narrator explains a concept is functionally a podcast with a face attached.

B-roll and stock footage fill time. They give the eye something to follow. Occasionally they add atmosphere. They almost never explain anything. A shot of a busy trading floor while you describe the 2008 financial crisis tells the viewer nothing about what actually happened.

Text overlays work for key terms, names, and figures. They are consistently overused as a substitute for content that would otherwise require real visual production.

Animated charts and maps carry explanatory information directly. A chart showing inflation rates across eight countries over five years communicates a relationship the viewer can read in two seconds. A map showing which European countries accepted refugees in 2015 and in what numbers makes the scale of the crisis tangible in a way that narration alone cannot. These visuals carry the argument.

Tools

Script: Use Claude or ChatGPT to get a first draft. Rewrite it in your own voice before recording. The draft gets you past the blank page. The rewrite is where the editorial judgment and the specific examples go in.

Narration: Record yourself or use a synthetic voice via ElevenLabs. Synthetic voice works if you choose a voice with natural cadence variation and edit it carefully. Flat, even delivery reads as generated. If every sentence lands with the same weight regardless of what it says, the narration sounds like output rather than someone explaining something.

Video editing: Premiere Pro, DaVinci Resolve, and CapCut all work. The editor is less important than the cut. Most explainer pacing problems are editing problems, not tool problems. Too many cuts, unnecessary B-roll padding, and silence between points all hurt retention more than the choice of NLE.

Animated graphics: After Effects is the professional standard and takes months to learn to a useful level. Moshion generates animated content from a text description: charts, maps, timelines, text animations, text highlights, and complex animated concepts. Export as MP4 and drop into any timeline.

Moshion generates animated content from a text description: charts, maps, timelines, text animations, text highlights, and complex animated concepts. Export as MP4 and drop into any editor.

Format decisions

Length: 8 to 15 minutes is where educational YouTube content tends to perform best on retention. Under 5 minutes rarely gives enough time to build and resolve a full argument. Over 20 minutes requires very high information density throughout to avoid drop-off.

Chapters: Add them to every video. They improve retention, appear as jump links in Google search results, and let a viewer who already knows part of the subject skip to what they need. A video without chapters on a complex topic is harder for new viewers to navigate and harder for search to surface.

Aspect ratio: 16:9 for the main video. If you want Shorts distribution, cut a separate vertical version. Auto-cropping a horizontal explainer for vertical destroys the visual composition, particularly for charts and maps that are built for widescreen framing.

Thumbnail: One clear visual claim or question. For faceless channels, a chart graphic or a specific data point as the thumbnail visual tends to perform better than face-plus-text combinations.

Where animated graphics fit in

For channels covering history, economics, geopolitics, or science, animated charts and maps are central to the format. The section of a video where a chart appears on screen while the narrator explains a relationship is often the section with the highest audience retention in analytics. Viewers rewatch it. A visual argument travels faster than a spoken one.

Custom animated maps built for a specific script show the viewer exactly what the narration describes. A map of Ottoman expansion at three points in time, animated to show the change, built specifically for your video on Ottoman history, is original visual production. Moshion generates it from a description in under a minute. The equivalent in After Effects would take a full day.

The output is rendered from exact specifications: defined colours, defined typography, defined motion paths. You can edit any element before exporting. The result looks like a motion designer built it by hand, because in every practical sense it is one.

A chart built for your script, showing the exact countries and time period you are discussing, is original production. It is made faster with AI, but it is built for that script and looks like it.