All Guides
Guides

How to Explain Complex Topics With Whiteboard Videos

Complex topics don't fail because your reader is dumb — they fail because prose demands the reader hold too many things in working memory at once. Whiteboard videos succeed for the opposite reason: the drawing sequence externalizes the reasoning. Here's how to use that on purpose.

Mira Chen17 min read
Bold editorial illustration of a dense tangled knot of equations, legal clauses, and molecular diagrams unwinding into a single clean line on a whiteboard — representing complex ideas being untangled into a whiteboard explainer video

Most complex topics don't fail because the reader is unmotivated, or lazy, or under-credentialed. They fail because prose asks the reader to hold too many things in working memory at once. A contract clause referencing three earlier definitions. A derivative pricing model that only makes sense once you already grasp its four variables. A clinical procedure whose second step assumes you've internalized the first. By paragraph two, half the scaffolding has quietly fallen out of the reader's head, and the writer, who never had to hold all of it simultaneously, has no idea it happened.

Whiteboard videos succeed at the same material for a reason that isn't obvious until you say it out loud: the drawing sequence externalizes the reasoning. Every earlier concept stays on screen. The metaphor persists. The narrator paces the reveal to match the drawing, so the viewer is never asked to hold more than one new idea at a time. It's not a stylistic preference. It's a working-memory hack. This guide is the practical version: where the format shines, how to structure a video that respects a smart audience, and the mistakes that flatten an otherwise good technical explainer.

TL;DR

  • Complex ideas fail as prose because working memory is finite; sentences push earlier context out of frame before the reader can integrate it.
  • Whiteboard videos work because the sketch stays on screen, the narration syncs with the reveal, and the viewer can pause and rewind.
  • Best fits: law, finance, healthcare, engineering, and science education. Any topic where the concept lives in the relationships between parts.
  • For real complex-topic work, Growth tier ($199.99/mo) is the honest recommendation. Color plus the 4-minute cap is what lets the reasoning actually unfold.
  • Try it first: pick one concept you've watched people politely nod through, chunk it into four to six atomic ideas, and generate a 1-minute preview.

→ Why the whiteboard format works  ·  Fields where this shines  ·  Five-step workflow  ·  Demo video  ·  Three prompt patterns  ·  How to chunk a complex topic  ·  Five common mistakes  ·  FAQ

Why the whiteboard format specifically works for complex ideas

You can explain a complex topic in many formats: a dense white paper, a dry lecture, a wall of Slack messages. None are wrong exactly, but they all fight the same enemy. Working memory is small, and complex material overflows it. Whiteboard video works because it addresses three bottlenecks at once.

1. Sequenced reveal. A whiteboard video draws the concept in the order the argument builds. Term A appears, gets defined, and stays visible. Term B appears next to it and gets connected. By the time the third concept lands, the first two are still on screen, anchoring the new material. Prose can't do this. Every sentence pushes the earlier ones out of frame. Static diagrams can't do it either, because they show the finished picture rather than the reasoning that produced it. The drawing, not the drawing's final state, is where the explanation lives.

2. Dual coding. Cognitive load research has spent forty years converging on a stubborn finding: information encoded through two channels (auditory and visual, in sync) is retained meaningfully better than through either alone. A whiteboard video is the cleanest applied case of dual coding you can build. The narrator speaks the concept; the sketch shows it. Neither is redundant; the two channels reinforce. You already know a diagram makes a hard idea easier and a good explanation makes a diagram click. The video braids both.

3. Pause and rewind. The most under-discussed advantage of video for complex material is control. Rereading a paragraph is friction: the reader has to find the sentence, hold their new context, and re-integrate. A video viewer scrubs backwards by three seconds. For a difficult topic, this is the difference between a viewer who gives up at minute two and one who lands the concept by minute three. The medium is patient with the audience in a way prose isn't.

Fields where this shines

Some categories of subject matter are almost purpose-built for whiteboard treatment. In each of these, the failure mode of prose is well documented, and the failure mode of whiteboard video is much less severe.

  • Law. Contract terms referencing prior clauses. Litigation strategy hinging on a discovery timeline. Regulatory frameworks with interlocking definitions. A whiteboard video keeps every piece on screen while the narrator walks through their relationships, exactly how a good lawyer sketches it on a legal pad for a client.
  • Finance. Derivatives, portfolio construction, discounted cash flow, tax structures. Each concept depends on the last, and each requires visual anchoring: a curve, a payoff diagram, a decision tree. Whiteboard is essentially the native format of finance instruction.
  • Healthcare. Procedure walk-throughs for clinicians. Patient education about a diagnosis or medication regimen. Public-health explainers for a lay audience. Anatomy paired with a calm narrator outperforms both text pamphlets and stock-photo carousels, and the trust register of a hand-drawn explainer suits patient-facing material better than slick corporate video.
  • Engineering and technical products. System architecture, API workflows, data pipelines, deployment topologies. Engineering already thinks in whiteboard diagrams. Onboarding new engineers, walking a customer through a technical product, explaining an incident to a non-technical stakeholder: all whiteboard-native.
  • Science education. Cellular processes, quantum analogies, ecological cycles, climate feedback loops. The subjects where the visual metaphor is the explanation. If you teach in any of these fields, Golpo for education covers the classroom-specific patterns in more depth.

The pattern: the concept lives in the relationships between parts, and those relationships need to be visible while the argument unfolds. If your topic fits that description, whiteboard video isn't a stylistic choice. It's the correct format.

The five-step workflow

The workflow below assumes you already have the topic and the source material. What it adds is the discipline of thinking through the video before generating it, which is the single biggest predictor of whether the output lands.

Step 1. Chunk the concept before you generate

The step most people skip and then regret. Before you write a prompt, sketch the video on paper. What are the atomic units? For a two-minute video, aim for four to six chunks; for four minutes, six to eight. Each chunk should be nameable in a sentence: "define the term," "show the mechanism," "give the example," "close with the takeaway." If you can't name a chunk in a sentence, it isn't one chunk. It's either two chunks pretending to be one, or a chunk you don't understand yet.

Step 2. Write an audience-specific brief

Complex topics fail when the writer forgets who's watching. State the audience explicitly: "A first-year medical resident, comfortable with anatomy but unfamiliar with the procedure. Assume they know terminology X but not concept Y." The specificity resets the AI's default register. Otherwise the script hedges both ways at once, over-explaining basics and under-explaining the novelty. One or two sentences of audience framing reshapes the whole output.

Step 3. Pick a single visual metaphor and commit

Complex topics need a spine. Choose one metaphor (a river through a system, a scaffold being built, a courtroom with two sides of a dispute, a signal traveling through a circuit) and instruct the tool to carry it across the whole video. The video_instructions field on Business tier earns its keep here: one sentence like "Use a river-and-tributaries metaphor throughout; the main concept is the river, each supporting idea is a tributary joining it" gives the video the through-line that separates a memorable explainer from a forgettable one.

Step 4. Generate with color and enough runway

Complex topics almost always need more than two minutes. Three to four is the sweet spot; shorter compresses the reasoning past usefulness. Color also does real work in technical explainers. Diagrams gain readability when you can differentiate arteries from veins, bull case from bear case, request path from response path. On Golpo that means Growth tier ($199.99/month, $166.66 annual): color plus the 4-minute cap. Business ($499.99/month) is worth the step up only if you're using video_instructions for consistent art direction across many videos, or if you need the API.

Step 5. Watch as the expert, then watch as the audience

Two passes. The first checks accuracy: did the narrator butcher a term, did the sketch miss a step, did the script skip a definition. The second pass is the important one. Pretend you're the target viewer and note every moment you'd have felt lost. Those are the places the video failed to chunk cleanly. Regenerate the segment, tighten the prompt, or split into two videos. Ten extra minutes; saves the video from mediocrity.

See it in action

The clearest way to understand why whiteboard works for complex material is to watch one. Below is a one-minute Golpo generation on a deliberately hard topic, the kind of subject that reads as a wall of jargon in text and lands cleanly in video.

A one-minute Golpo demo showing how a complex concept becomes a whiteboard explainer.

Notice what the video does that a same-length paragraph couldn't: earlier terms stay visible as later ones connect to them, the narrator's pacing matches the drawing so the viewer never holds two new ideas at once, and one visual metaphor carries the argument. That last piece is what most people miss on a first attempt.

Three prompt patterns that reliably work

You can write your own from scratch, but these three cover most of what technical creators need. Copy, adapt, paste your source material. If your material lives in a PDF, our convert PDF to AI video guide walks through the upload workflow.

Pattern 1. Break down a technical concept

"Explain [CONCEPT] in a three-minute whiteboard video. Audience: [SPECIFIC AUDIENCE, e.g., 'a first-year law associate who hasn't seen this doctrine before but is comfortable with contract basics']. Structure: define the term in the first 20 seconds, walk through the underlying mechanism in the middle, then close with one concrete example and one common misunderstanding. Use one running visual metaphor throughout. Source material below."

Pattern 2. Compare two frameworks

"Compare [FRAMEWORK A] and [FRAMEWORK B] in a three-minute whiteboard video. Audience: [SPECIFIC AUDIENCE]. Show them side by side on the canvas so the differences are visible, not just described. Cover: what problem each was designed to solve, where their assumptions differ, and one situation where each is the right choice. End with a decision rule the viewer can use next time they encounter the choice."

Pattern 3. Walk through a specific case

"Walk through [SPECIFIC CASE / SCENARIO / PROCEDURE] step by step in a three- to four-minute whiteboard video. Audience: [SPECIFIC AUDIENCE]. For each step, show what's happening on the canvas and explain why that step is necessary. End with the two most common places where this process goes wrong in practice. Source detail below."

The third pattern is the one that translates deepest expertise into video most reliably. If you have a scenario you've walked colleagues through a dozen times, drop the notes in and generate. If your notes live somewhere already, the notes-to-video workflow is worth reading first. The video captures the walkthrough at a scale you can't personally.

How to chunk a complex topic before generating

The single highest-leverage habit for anyone making these videos is chunking discipline. The rules are short but non-negotiable:

  • One idea per screen. If a chunk contains two concepts, split it. The format only works if each visual anchor corresponds to a single, named idea.
  • Name the concept before you show it. Narration introduces the term; the sketch then draws it. Naming primes recognition; unnamed visuals spend working memory on the wrong thing.
  • Use one visual metaphor and use it well. A river, a scaffold, a courtroom, a signal path. Introduce in the first 20 seconds and reinforce throughout. Two metaphors is one too many.
  • Resolve the metaphor by the end. The satisfying moment in a good explainer is when the metaphor closes: the river reaches the sea, the scaffold reveals the finished building. Closure is what makes the video memorable rather than merely informative.
  • Close with a takeaway the viewer can act on. Not a summary. A single sentence they could say out loud tomorrow. "The three-part test is X, Y, Z." "Discount cash flows, not earnings." "If the patient reports X, escalate immediately."

If you can't do all five for a topic, you're not ready to generate yet. Sit with the outline for ten more minutes; it's cheaper than regenerating three times.

Five common mistakes that flatten a technical explainer

  1. Cramming the whole topic into two minutes. Complex material has irreducible length. Two minutes forces the AI to skip the connective tissue between concepts, which is where understanding lives. Growth tier's four-minute cap exists precisely for this.
  2. Skipping the audience specification. "Explain compound interest" produces a generic explainer. "Explain compound interest for a small-business owner deciding whether to prepay a loan" produces a video that lands. Specificity is the whole game.
  3. Using two visual metaphors. Once you introduce a metaphor, don't switch. Switching mid-video costs the viewer the anchor and they don't recover.
  4. Assuming the narrator preserves technical accuracy on its own. AI voices occasionally mispronounce terminology. Watch once with the transcript open; verify critical terms and re-render if wrong.
  5. Writing for the expert reviewer rather than the actual viewer. The temptation is to satisfy the smartest person who'll see it, usually a colleague, not the target audience. The video should land for the audience even if the colleague finds it slightly under-detailed.

Try it on a topic you've been struggling to explain

Open Golpo and pick one concept you've written about, presented, or explained in meetings and watched people politely nod through. The "Just exploring" free preview generates a one-minute watermarked sample: enough to confirm the format fits your topic. When you're ready to ship, Starter ($39.99/month, $33.33 annual) covers downloads and 2-minute B&W videos. Creator ($99.99/month, $83.33 annual) unlocks voice_instructions, multilingual, and vertical output. Growth ($199.99/month, $166.66 annual) is the sweet spot for complex-topic work with color rendering plus the four-minute length cap. Business ($499.99/month) adds video_instructions, voice cloning, and API access for teams producing explainers at scale.

Frequently asked questions

Is Golpo accurate enough for technical or legal content?

It's accurate enough that we use it, but not accurate enough to skip a review pass. Golpo works off your source material, so the terminology and facts come from what you paste in. That said: always watch once with the transcript open before shipping. The narrator can mispronounce a term of art, and the sketch occasionally simplifies a distinction you care about. For a case brief or a clinical protocol, treat the first render as a strong draft, not a finished product.

How do I keep diagrams precise instead of vaguely "illustrative"?

Name what you want in the source material. If the diagram should show four labeled components in a specific arrangement, write that out: "Draw a box labeled A on the left, connected by an arrow to a box labeled B on the right." On Business tier, video_instructions lets you pin the exact visual through-line. Below Business, the trick is being explicit in the script itself. Vague prompts produce vague sketches; specific prompts produce specific ones.

Can I feed it a research paper or a case brief?

Yes. PDF upload is the intended path for long-form source material. See our convert PDF to AI video guide for the workflow. The main thing to know: Golpo will happily condense a 40-page paper into three minutes, and the condensation is exactly where you need to steer it. Add a one-paragraph brief telling it which section is the argument, which is background, and what the takeaway should be. Otherwise it may spend a minute on the literature review.

Does the AI oversimplify?

By default, yes. That's what AI models trained on the open web do when they see a hard topic: they reach for the friendliest analogy and hedge. The fix is in your brief. State the audience's expertise level explicitly ("a first-year law associate comfortable with contract basics") and name one or two concepts you do not want simplified. Once the model knows the register, it holds the level much better. The voice_instructions field on Creator+ helps here too, because the register of the narration reinforces the register of the writing.

What about equations and formal notation?

Whiteboard videos handle equations well in the sense that the sketch can write them out step by step, which is often better than a static rendered equation. The narrator talks through each symbol as it appears, so the viewer never has to parse the whole line at once. Where it struggles: dense mathematical typography (fractions inside integrals inside subscripts). For those, either break the equation into stages the sketch draws sequentially, or pair the video with a supplementary image of the fully typeset formula.

Can I control what appears in each frame?

You control the script, and the sketch follows the script. So if your script says "now draw a decision tree with three branches labeled A, B, and C," that's what appears. Frame-by-frame direction (this exact pixel, this exact position) isn't exposed, and honestly you don't want it: the strength of the format is that the AI paces the reveal to the narration. What you do want to control is the sequence and the labels, and both live in the script.

Is 2 minutes enough for a complex idea?

Usually no. Two minutes is fine for a single concept with two supporting ideas. Anything with more moving parts (a doctrine with three prongs, a pricing model with four inputs, a procedure with five steps) needs three to four minutes to breathe. That's the honest reason we point complex-topic work at Growth tier ($199.99/month): the 4-minute cap is the thing that lets the reasoning actually unfold instead of getting compressed.

Tags

#Explainer Videos#Whiteboard#Technical Content#2026

Keep reading