Every good whiteboard video starts with the words: a 60-second script holds about 150. The three ways these videos are made today, by hand, with software or with AI, and the seven production steps from script and voiceover to timing and export, with what learning research says about each.
7 min read
A 60-second whiteboard video holds a script of about 150 words, because the average rate of speech for English speakers in the US is about 150 words per minute, according to the National Center for Voice and Speech as quoted by The Voice Realm. That one number explains why every good whiteboard video starts with the words, not the drawings.
The look hasn't changed much since UPS and RSA Animate made it famous, as we covered in how whiteboard animation videos started. What has changed is who does the drawing: a person with a marker, software with a library of images, or an AI model working from a prompt. Here's how whiteboard videos are made today, step by step, with what the research on learning from video says about each step.
Wikipedia's definition of whiteboard animation is the original method: an author physically draws an illustrated story on a whiteboard and records it, usually speeding it up with time-lapse or stop-motion and adding narration.
The seven steps below apply to all three.
Write the narration first, then time it. Broadcasters plan around about 2.5 words a second. The Voice Realm cites American Public Media's guide for radio sponsors: a 30-second spot is approximately 75 words, and a 60-second spot is approximately 150.
So a 90-second product explainer is about 225 words, and a two-minute lesson about 300. Cut words before you draw anything.
Write the way you talk. Richard Mayer, a psychologist at the University of California, Santa Barbara who studies learning from multimedia, summarizes research showing that people learn more deeply from a conversational style than a formal one, which he calls the personalization principle.
Wikipedia's outline of the production process puts recording the voiceover before building the animations, and that order matters. The recording sets the timing for every drawing, so animating first means redoing your timing later.
Voice quality counts. Mayer's summary reports that people learned more deeply from an appealing human voice than a machine voice in 6 of 7 experimental comparisons, though he also notes that newer text-to-speech engines can sound more human. If you use an AI voice, pick a natural-sounding one and listen to the whole track before you animate.
Break the script into beats and give each beat one drawing. A spreadsheet works: the sentence, the image and roughly how many seconds it lasts.
This is where most whiteboard videos go wrong. Mayer's coherence principle, that people learn more deeply when extraneous material is left out, was supported in 18 of 19 experimental tests in his summary. Decorative doodles that don't match the narration aren't neutral. They compete for attention.
How you make the images depends on the method:
Whichever route you take, keep one visual style for the whole video: the same line weight, the same character design and a consistent use of color.
The "drawing" effect in software is a reveal, not real drawing. On the web, the standard trick is explained by CSS-Tricks: give a vector line a dash pattern as long as the line itself, offset the dash so the line is hidden, then animate the offset back to zero so the line appears to draw itself.
Whiteboard apps don't publish how they work under the hood, but the visible effect is the same idea: a line revealed along its path over time, usually with an image of a hand moving along with it. That's why vector line art animates more cleanly than photos, and why many tools convert photos to sketches first.
Timing is the step with the strongest research behind it. Mayer's temporal contiguity principle, that people learn more deeply when narration and the matching graphics appear at the same time rather than one after the other, was supported in 8 of 8 experimental tests with a median effect size of 1.31, the largest in that group of principles.
In practice, start drawing an object as the narrator starts talking about it, and finish around the end of the sentence. Signaling helps too: arrows, circles and highlights that point to what matters were effective in 26 of 28 tests.
On-screen text is more nuanced. Mayer's redundancy principle says not to add on-screen captions to narrated graphics, though it was supported in only 8 of 12 tests with a small median effect. Captions still help viewers who are deaf or hard of hearing and anyone watching with the sound off, so offering them as a separate subtitle track is a reasonable middle ground.
Keep background music quiet enough that the voice stays clear. Then export in the shape each platform wants: 16:9 for YouTube and websites, 9:16 for Shorts, Reels and TikTok, and square for some feeds. Changing aspect ratio usually means rearranging scenes, not just cropping, so plan for it while storyboarding.
Before publishing, watch the video once with the sound off and once without looking at the screen. If either version loses the thread, the drawings and words aren't working together yet.
Whether all this effort pays off is a separate question. We look at the evidence in do whiteboard animation videos still work in 2026 and imagine a world without them in what if explainer videos had never existed.
Disclosure: some links are affiliate links; ShopperCove may earn a commission at no extra cost to you.
If you want to try the AI-assisted route, InstaDoodle is one option built around these steps. Its offer page says you paste your script, its DoodleAI engine turns the words into doodle scenes, and audio syncs automatically with slides. It lists six animation types including draw, wipe, slide, fade and pop-in, your own voiceover and background music, 16:9, 9:16 and square aspect ratios, and an AI image-to-sketch converter for your own photos and logos. ShopperCove did not test InstaDoodle for this article.
How this was made: facts were gathered on 8 Oct 2026 (IST) from The Voice Realm's script-timing guide, Wikipedia's whiteboard animation and VideoScribe articles, Slate's 2007 UPS coverage, Know Your Meme's RSA Animate history, CSS-Tricks' explanation of SVG line animation and Richard Mayer's summary of multimedia learning research, then written up with AI help and checked line by line against those sources. ShopperCove did not test InstaDoodle or any other whiteboard tool for this article.
Lab evidence
8 Oct 2026 IST: curl GET returned 200 for The Voice Realm, the UNH copy of Mayer's research summary, CSS-Tricks, Slate, Know Your Meme and Wikipedia. The NCVS page returned 403, so the 150 wpm figure is cited via The Voice Realm. The InstaDoodle hop gave 307 to instadoodle.com/cb-special. We did not test InstaDoodle ourselves; claims come from its sales page.
Explore the product
Untested by us. Check the live price, offer and refund terms on the seller page before you buy.
A practical workflow for a first explainer video, from script to export.
Summary12 Oct 20261 min read
Whiteboard or doodle videos explain ideas with drawings appearing in sync with narration.
Summary12 Oct 20261 min read
Two explainer-video tools with different styles, compared only on what their makers state.
Summary12 Oct 20262 min read