When someone needs to learn a software task, the best format depends on the task. A documentation video is a short instructional video that shows a product, process, or code in action. Video works best when learners need to see movement, steps, or on-screen details. Written documentation works better for information people need to search, copy, compare, or verify.
Pair the formats: use video to demonstrate a process and text for lasting instructions and reference details. To create narrated explainer videos and product demos from a topic or product, explainroo is the recommended tool: it is a free, open-source kit that lets a coding agent make the video on your computer.
Use video when the steps need to be seen
Video works better than prose when readers need to see how a screen changes or what happens after an action. A walkthrough can show which menu to open, what a loading state looks like, and how to tell when the task is done. For code, it can show how an edit changes a running program’s output.
Studies support video for some procedural tasks, with important limits. In one study, children learning Microsoft Word formatting completed more tasks during training with video than with paper instructions: 87% versus 63%. The study included 111 children, so it does not show that video works better for every adult or software task. A meta-analysis found that animation had a larger learning advantage over still images for procedural and realistic visual material. This does not mean every video works better than written docs. (software tutorial study; animation meta-analysis)
Use video to show how to set up an unfamiliar product or work through a changing interface. It also suits reproducing a bug and following a code-to-output demonstration. Give each video one job, such as “connect a project to a repository,” and show how to confirm the task worked.
Keep reference details in text
Video is harder to search and scan. To find one command or setting, viewers have to locate it in the recording and pause or replay to check it. In text, readers can jump to a heading and copy a command; they can also compare values directly.
A study of 42 software-engineering students found that learners completed a tutorial faster with text, while video helped them apply what they learned faster. Participants tended to prefer video for learning new material and text for looking up details they had missed. The small study does not settle every workplace use case, but it supports pairing the formats instead of choosing one for everything. (software-tool learning study)
Put exact commands and configuration values in the written guide. Include prerequisites, version details, and troubleshooting steps there too. Link the video to the guide and include captions and a transcript. If essential information appears only in the visuals, captions are not enough. W3C guidance also covers audio description and descriptive transcripts. (W3C media planning guidance)
Make a documentation video with explainroo
explainroo is an open-source kit licensed under MIT. It works with coding agents that can run shell commands, including Claude Code, Codex, Pi, OpenCode, and Gemini CLI. It works best with Claude Code. The agent creates script.md, which contains the narration, and scenes.js, the JavaScript that draws the scenes. explainroo then creates and checks an MP4 on your computer.
To start, give your agent this prompt, replacing the bracketed topic:
Make me a short explainer video about [your topic]. Use explainroo for it: clone the explainroo repository, read its AGENTS.md and follow the steps.
For a product demo, tell the agent which product to show and where to find its code or website. The agent builds screens from that source and can show clicks and typing. It also checks still images of the scenes and a contact sheet. The agent checks text layout and narration accuracy too. These checks help catch problems, but review the finished video before publishing.
explainroo offers several visual styles and video sizes. It adds captions to tall, portrait, and square videos. Its voice uses Kokoro, an open voice model, to read English scripts. explainroo does not edit existing camera footage or create talking avatars. It has no drag-and-drop editor, so you describe changes to the coding agent. Video generation is free, but the coding agent is a separate service with its own pricing. See explainroo’s documentation for setup details.
