TL;DR
- The first decision for an explainer isn't the script. It's what the explainer is made of, because an animated, a filmed and an AI-presenter explainer are three different productions, and three quotes for "an explainer" are rarely for the same thing.
- Animation shows what a camera can't: the inside of a machine, a process side by side, a product that isn't built yet. Live action proves it's real. An AI presenter suits the same explanation said many times, in several languages or versions.
- For English, BM and Chinese versions, animation and voiceover-led footage change the voice and the on-screen text; a presenter filmed speaking to camera is the hardest to turn into another language.
- If the explainer shows prices, specs or an interface, keep those parts in a format that can be edited. Use the camera for what won't change.
- A mix often works best for B2B products: open on real footage to show the company and product are real, then animate the part that has to be understood.
The product launches next quarter, and the sales team keeps losing people at the same slide. Someone finally says it: we need an explainer.
Most teams go straight to the script, or straight to asking for quotes. Both skip a decision that shapes everything after it: what the explainer is going to be made of. An animated explainer, a filmed one and one delivered by an AI presenter can carry exactly the same script, but they are three different productions. They are made differently, they age differently, they handle languages differently, and they ask the viewer to believe different things. Ask three studios for "an explainer" without settling this, and you can end up with three quotes for three different products.
So decide the medium first. Here's how to think about it.
What's the difference between an animated, a live action and an AI presenter explainer?
Each one is good at a different job.
Animation draws what a camera can't film. The inside of a machine. Money moving between three accounts. A six-step process laid out side by side. A product that hasn't been manufactured yet. Nothing on screen has to exist first, and nothing is on screen unless it helps the point, which is why so many explainers are animated. The limit is the other side of the same strength: everyone knows anything can be drawn, so an animation can't prove that something is real.
Live action proves it exists. A real factory floor, a real team, the product working in a real customer's hands. When the viewer's doubt is "is this real, and can these people actually deliver it?", footage answers that in a way a drawing can't. The limit is that a camera only shows what's in front of it. If the thing that needs explaining happens inside software, inside a pipe or inside a contract, filming the outside of it doesn't explain much. And if someone explains it to camera, who that is matters. A clear presenter often does better than your most senior engineer, for reasons we've gone through in who should be on camera.
An AI presenter is for the same explanation said many times. A generated presenter reads a script you approve. It suits explainers that need many versions (several languages, several products in a range, one per market) or that will be revised often. The trade-off is trust. An AI presenter costs some credibility with the viewer, which is acceptable for "here's how the feature works" and wrong for "here's why you should trust us with your money". Where it fits and where it doesn't is covered in AI presenters and digital spokespeople, so we won't repeat it here.
Put shortly: animation is for understanding, live action is for belief, and an AI presenter is for repetition.
How do you make one explainer in English, BM and Chinese?
In Malaysia this is often the deciding question, and the three handle it very differently.
- Animated. The pictures usually stay as they are. The voiceover is recorded again for each language, and any words on screen are swapped. That second part is easy to underestimate: a label that fits neatly in English can run noticeably longer in Bahasa Malaysia, and Chinese text sits differently on screen. On-screen text works best when it's planned for every language from the start, not squeezed in afterwards.
- Live action. It depends on whether anyone speaks to camera. If the footage runs under a voiceover, the voiceover can be replaced language by language, much like animation. If a presenter speaks on camera, the choices are subtitles, a dub over them, or filming them again in each language. Subtitles ask more of the viewer, a dub over a visible speaker shows because the lips don't match, and filming again repeats the shoot.
- AI presenter. Changing language is what the format is for: the same presenter delivers the script in each language. What doesn't change is that someone fluent has to check each version before it goes out. A generated line can be grammatically correct and still sound wrong to a native speaker.
If you know at the start that you'll need more than one language, say so in the brief. It changes how the explainer should be made, not just how it's finished.
How we usually do it: subtitles in two or three languages under one main voiceover. That covers most briefs, and it keeps the cut and the voice the same across every version. When a client wants a separate voiceover for each language, we do that too, for animated and filmed explainers alike, but every extra voiceover is new hours of work, so it's quoted on top of the main version. AI presenters we've delivered in all three languages.
Which kind of explainer is cheapest to update later?
Before choosing, ask how long the explainer will stay correct. If it shows prices, specifications, a user interface or a regulation, the answer is probably "not long".
- Animated. A changed number or a redesigned screen usually means changing the scenes it appears in, not remaking the whole piece. How cheap that is depends on two things: the working files being kept, and the change being small. A new product shape that appears in every scene is closer to a remake.
- AI presenter. The cheapest to revise. A changed line is a script edit, and the section it affects is regenerated. With us, a change like that is quoted as a small job on its own, so it's worth batching edits that arrive close together rather than sending them one at a time.
- Live action. The hardest to update. A changed fact in a filmed shot means going back to the location, the people and the equipment, or cutting the shot out. That's fine for what doesn't change, like your facility, your team and why you exist, and expensive for anything that does.
The practical rule: put the parts that will change in a format that can be edited, and keep the camera for the parts that won't.
How it works with us: we keep the project files for 90 days after delivery, the same window we keep raw footage and masters. Inside that window, a changed price or spec means editing the scenes it appears in. After it, the files are gone and an update is close to a remake. If an explainer will need updating for years, the answer is to keep the files on your side: bring a drive or your own storage and we hand the source files over. Whether that handover is charged depends on the project and how long we've worked together, so it's part of the quote. If you'd rather we keep the files for you, that's storage we look after, and it's charged for every client.
Can you film part of it and animate the rest?
Yes, and for a lot of B2B products it's the structure we'd suggest first: open on live action to establish that the company and the product are real, then switch to animation for the part that has to be understood, such as how it works inside or how the steps fit together. If the viewer needs to remember who they're dealing with, close on real footage again.
This follows the same rule as getting more than one piece out of a single shoot: decide what leads, and plan the rest around it before the cameras roll. The filmed half has to be shot knowing where the animation will take over, or the cut between them looks like two videos joined together.
One caution carries over from who should be on camera: mix the formats between sections, not within a single moment. A real customer introduced by an AI presenter feels slightly off to viewers, even if none of them could say why.
When shouldn't you film your explainer?
Filming is usually the biggest spend of the three, and it's the work a production studio is best known for, so this is worth saying plainly. We'd advise against filming an explainer when:
- The product doesn't exist yet, or only as a prototype that will still change. You'd be filming something you'll have to film again. It's the same problem we worked through for property launches in what a trailer can generate and what has to be filmed.
- What needs explaining can't be seen. Software logic, a financial product, a process sealed inside a machine. Footage of people at laptops or equipment humming in a room doesn't explain any of it.
- It will change within the year. Pricing, plans, interfaces, rules.
- It's led by a presenter and you need it in three languages. That's three shoots of the same thing.
It works the other way round too. Don't animate when the viewer's question is "is this real?" If a buyer needs to see that your plant exists, that your team is who you say it is, or that the product works in the field, a drawing doesn't answer them. And don't use an AI presenter when the viewer has to trust the person speaking.
Sometimes the answer is none of the three. If the explainer is really "how to use our software", a clear screen recording with a good voiceover usually teaches better than any of them.
So which one does your explainer need?
Run your explainer through four questions: is there something real to film, how many languages does it need, how often will it change, and does the viewer need to trust it or to understand it?
| Your explainer is about | Something real to film? | Languages | Changes | Trust or understanding? | Usually best made as |
|---|---|---|---|---|---|
| A software or app feature | No, it happens on a screen | Any | Often | Understanding | Animated; a screen recording if it's "how to use it" |
| How a financial or insurance product works | No | Often two or three | When the terms change | Understanding | Animated, or an AI presenter if you need many versions |
| An industrial product or a plant, for buyers who need to see you can deliver | Yes | One or two | Rarely | Trust | Live action, with animation for what happens inside |
| A product that isn't manufactured yet | Not yet | Any | Will change before launch | Understanding | Animated |
| The same explanation across many products, markets or languages | Sometimes | Several | Often | Understanding | AI presenter |
| A process customers keep getting wrong | Sometimes | Several | When the process changes | Understanding | Animated; filmed if they have to recognise a hands-on step |
Two things can change the answer:
- If nobody on your side will own the updates, the formats that are cheap to edit lose their main advantage. Decide who signs off a change before you choose.
- If the explainer is the first thing a stranger will see of your company, it's carrying more trust than an explainer usually does. That tips the opening towards live action, even if the rest is animated.
What this page can't settle
It doesn't compare prices. The three are priced on different things, which is exactly why three quotes for "an explainer" are rarely comparable until the format is decided. Decide the format, then compare the quotes.
It also can't write the explanation for you. What your product does, and whether each claim in the script is accurate, is yours to sign off. We can make it clear. We can't make it correct if it wasn't to begin with.
If you're working this out for a launch, tell us what you need to explain and how many languages it has to be in. Those two answers settle most of the decision.
Thinking about a project like this?
Tell us what it needs to do — we'll take it from there.
Frequently asked questions
- Should an explainer video be animated or live action?
- It depends on what the viewer needs. If they need to understand something a camera can't show, like how software works inside, how money moves, or a product that isn't built yet, animation does that better. If they need to believe the company and the product are real, live action does that better. Many B2B explainers use both: real footage to open, animation for the mechanism.
- When is an AI presenter a good choice for an explainer?
- When the same explanation has to exist in many versions, such as several languages, several products in a range, or frequent revisions. A changed line becomes a script edit rather than a reshoot. The trade-off is some loss of trust with the viewer, so it suits "how this works" better than "why you should trust us".
- Which kind of explainer is easiest to update later?
- An AI-presenter explainer is the easiest, because a change is a script edit and the affected section is regenerated. An animated explainer can usually be updated scene by scene if the working files are kept. A filmed explainer is the hardest, because a changed fact in a shot means going back to the location and the people, or cutting the shot.
Keep reading
- Training Videos for Your Staff: Film Them, Use an AI Presenter, or Screen-Record?Before you decide who makes your onboarding or SOP videos, decide how often they'll have to change. What a proper shoot is worth spending on, where an AI presenter does the job, when a screen recording is the whole answer, and how to mix the three.
- What Is a Brand Film (and When a Corporate Video Isn't Enough)?A brand film goes after belief, not information. Here's what actually makes one, how it differs from a corporate video, and the moment a straight capability video stops being enough.
- Corporate Video, Brand Film, TVC, Explainer: Which One Do You Actually Need?Four kinds of company video get lumped together and briefed as if they're interchangeable. They're not. Here's what each one is for, and how to pick the right one before you spend.