The Best 2D Motion Graphics Videos — and the Research Behind Why They Work

Every list of the best 2D motion graphics videos is a list of things that look nice. That is a legitimate way to enjoy the craft and a useless way to commission it. This piece takes the opposite approach: the seven properties that separate motion graphics which transfer understanding from motion graphics which merely move — each one with the research that supports it.

The only criterion that matters

A motion graphics video is good if a person who did not previously understand the thing now understands it, and can still explain it an hour later. Everything else — the easing, the palette, the typography, the sound design — is instrumental. It is either serving that outcome or it is decoration.

This is not an anti-craft position. Craft is precisely how the outcome is achieved. It is an argument about what the craft is for, and it has consequences for what you should be looking at in a showreel.

d = 1.06 the effect size favouring animation over static pictures when the subject is procedural knowledge — something with steps, done in an order. Across all subject types the advantage was a more modest d = 0.37 (26 studies, 76 comparisons). Höffler & Leutner (2007), Learning and Instruction, 17(6), 722–738[1]

That gap between 0.37 and 1.06 is the whole discipline in two numbers. Motion is not generically better than stillness. Motion is better at showing change. The best 2D motion graphics videos are, almost without exception, videos about something that changes — a process, a flow, a transformation, a before and after. The weakest are animated versions of things that were never moving in the first place.

Seven properties of 2D motion graphics that work

An overview of 29 systematic reviews, covering 1,189 studies and 78,177 participants, identified eleven multimedia design principles with significant positive effects on learning.[2] Seven of them translate directly into things you can see in a finished piece.

1. One thing moves at a time

Signalling — directing attention to the relevant part — was among the largest benefits identified. In motion graphics this means restraint: if six elements animate simultaneously, none of them is emphasised, and the viewer chooses where to look at random.

Watch any piece you admire with the sound off and count how many things move in each beat. Usually one. Occasionally two, in a relationship. The busy-ness you remember is the sum across time, not the density at any moment.

2. It narrates the argument

The modality principle holds that spoken explanation with visuals outperforms the same words printed on screen. The failure mode is everywhere: a voiceover saying one sentence while a different sentence is typed on screen, and the viewer stops listening to read it.

Text belongs on screen for labels, numbers and the things worth remembering exactly. Not for the argument.

3. It cuts the charming bit

The coherence principle, and the evidence on seductive details, is unsentimental: interesting but irrelevant material measurably reduces learning. The sequence everyone in the review loved, that has nothing to do with the point, pulls the viewer’s attention away from the explanation, and comprehension drops with it.

4. It is segmented, and the viewer controls the pace

Segmenting has robust support. This is why a 90-second explainer plus three 20-second cutdowns frequently outperforms one three-minute film, and why the best product motion graphics are increasingly built as a set: one master and its cutdowns.

5. It sounds like a person

Personalisation — a conversational register — is one of the principles with consistent positive effects. The study of 6.9 million video sessions found the same thing from a different direction: informal, human-feeling delivery outperformed highly produced studio material.[3]

“Your team wastes two hours a week on this” outperforms “Organisations experience significant productivity leakage”, and it takes exactly the same effort to animate.

6. Words and pictures arrive together

Temporal and spatial contiguity — putting related words and images in the same place at the same moment — was among the largest effects in the overview. In practice: the label appears when the thing appears, and next to it, not in a caption three seconds later.

7. It is short

6 min the point past which engagement fell away sharply across 6.9 million video watching sessions on four edX courses. For commercial motion graphics the practical ceiling is far lower — 60 to 90 seconds for an explainer, 15 to 30 for social. Guo, Kim & Rubin (2014), Proceedings of ACM Learning @ Scale[3]

Three failure modes that look like quality

These are the ones that survive review meetings precisely because they look polished.

The transition showreel. Every scene connects with an inventive morph. It is genuinely difficult to execute and it is the most common way a piece stops teaching: the viewer’s eye goes to the transitions, and each one takes a beat of attention that the argument needed.

The isometric city. A beautifully built world where tiny figures move between buildings while a voiceover explains a payments API. The world is irrelevant to the explanation — the coherence principle in its purest violation — and it is usually the most labour-intensive part of the production.

The kinetic type wall. Every word of the voiceover appears on screen, choreographed. It looks like effort and it is the modality principle inverted: the viewer now has to read and listen to the same content simultaneously, which is harder than either alone.

How to judge a reel in ninety seconds

Four tests, in order. They take about as long as the reel does.

  1. Mute it. Can you still follow roughly what the product does? If the pictures carry no argument without the voice, you are buying decoration for a script.
  2. Count what moves. Per beat. If it is consistently more than two, the piece has no hierarchy.
  3. Wait an hour and try to explain the product to someone. This is the actual test and almost nobody runs it.
  4. Ask what the piece is about. Not what it is for — what it is about. A strong studio answers with a single idea. If it answers with the client’s feature list, the reel is a portfolio of technique.

Need motion graphics that survive the mute test?

We build the argument first and animate second. Tell us what you are explaining and who keeps misunderstanding it.

Start a project →

Writing a brief that produces the good version

Motion graphics briefs usually specify look and length. The good ones specify comprehension. Six lines:

  1. The one sentence. What should the viewer be able to say afterwards? One sentence, in their words, not yours.
  2. What they currently get wrong. The specific misunderstanding. This is the most useful line in any animation brief and the one most often missing.
  3. What changes over time. Name the process, flow or transformation. If nothing changes, question whether this needs to move at all.[1]
  4. What must be exact. Numbers, names, regulated claims, anything legal has to sign off. Easy to flag now, slow to fix after animation.
  5. Where it plays, and whether there is sound. Muted autoplay in a feed is a different design problem from a narrated piece on a landing page.
  6. The deliverable list. Master, cutdowns, ratios, subtitle files for each language you run in across APAC. Counted.

A studio that receives that brief and comes back asking about the second line is the one to hire. A studio that comes back with three style frames has answered a question you did not ask.

The short version

The best 2D motion graphics videos are about something that changes, move one thing at a time, narrate the argument, cut the charming bit, arrive in segments, sound like a person, and finish early. None of that is about taste. All of it has effect sizes attached. Judge a reel with the sound off, and build the brief around the misunderstanding.

Frequently asked questions

What makes a good 2D motion graphics video?

A good 2D motion graphics video leaves a viewer who did not understand the thing able to explain it an hour later. The strongest pieces are about something that changes, move one thing at a time, narrate the argument, cut irrelevant detail, arrive in segments, sound like a person and finish early.

How long should a motion graphics explainer video be?

Aim for 60 to 90 seconds for an explainer and 15 to 30 seconds for social. Across 6.9 million video watching sessions, engagement fell away sharply past roughly six minutes, and commercial motion graphics sit far below that ceiling. Segmenting also helps: a 90-second explainer plus three 20-second cutdowns frequently outperforms one three-minute film.

How do I judge a motion graphics studio's showreel?

Run four tests in about ninety seconds. Mute the reel and check you can still follow what the product does. Count what moves in each beat; more than two means no hierarchy. Wait an hour and try to explain the product to someone. Then ask the studio what the piece is about, and listen for a single idea.

Can motion graphics videos be localised for audiences across APAC?

Yes. Motion graphics are built from type, shapes and voiceover, so one master can be re-versioned for markets such as Malaysia or Indonesia with a new voiceover and translated labels. Plan it at brief stage: list every language in the deliverable count, and keep on-screen text to labels and numbers so each version stays simple to update.

References

  1. Höffler, T. N., & Leutner, D. (2007). Instructional animation versus static pictures: A meta-analysis. Learning and Instruction, 17(6), 722–738. doi:10.1016/j.learninstruc.2007.09.013
  2. Noetel, M., Griffith, S., Delaney, O., Harris, N. R., Sanders, T., Parker, P., del Pozo Cruz, B., & Lonsdale, C. (2022). Multimedia design for learning: An overview of reviews with meta-meta-analysis. Review of Educational Research, 92(3), 413–454. doi:10.3102/00346543211052329
  3. Guo, P. J., Kim, J., & Rubin, R. (2014). How video production affects student engagement: An empirical study of MOOC videos. Proceedings of the First ACM Conference on Learning @ Scale, 41–50. doi:10.1145/2556325.2566239

Read next