To make a seminar video presentation in 30 minutes, spend the first 10 minutes writing, not designing: one sentence stating the idea, four or five short paragraphs that each explain one part, and two discussion questions to finish. That is about 600 to 750 words, a five-minute video at 150 spoken words a minute. Spend the next 10 minutes drafting it into narrated scenes with a drawn visual per paragraph, and the last 10 checking names, dates and terms, then exporting. A seminar presentation is a starter for discussion, not a lecture, so the questions at the end matter more than the polish in the middle.

What a seminar presentation is for

A seminar is not a lecture with a smaller audience. The presenter's job is to put a reading or a model in front of the group clearly enough that the next 40 minutes can be spent arguing about it. That changes what a good presentation looks like.

It summarises, then it opens. Two thirds of the time explains the idea. The last third points at where it is weak, contested, or interesting.

It is short. Most seminar slots for a student presentation are five to ten minutes. Five is plenty for one model or one reading, and a tutor will rarely complain that you left more time for discussion.

It is accurate about the source. The group has usually done the same reading. Getting the author's name, the date or a key term wrong is noticed immediately.

That last point is why the 30 minutes below are not split evenly between writing and making. The writing carries the presentation, and the checking protects it.

The 30 minutes

Two rules keep this inside half an hour. Do not open any video tool until the script exists, because writing inside an editor is how 30 minutes becomes two hours. And do not reread the whole source. You did the reading already; skim only the parts you are unsure of when you reach them in the script.

Worked example: the sender-receiver model

A typical first-year seminar topic in communication studies, media, linguistics or business. The script below is about 260 words, a little under two minutes. A full five-minute version adds a paragraph on each later model and some examples from the reading.

The one sentence: "The sender-receiver model says communication is a message moving from a source to a destination through a channel, and that noise along the way can change what arrives."

The script:

In 1948, Claude Shannon, an engineer at Bell Labs, published a paper called A Mathematical Theory of Communication. A year later, with Warren Weaver, it became a book. Their model was built for telephone lines, but it became one of the most quoted diagrams in communication studies.

The model has a straight line of parts. An information source produces a message. A transmitter turns it into a signal. The signal travels along a channel. A receiver turns the signal back into a message. And the message arrives at its destination.

Above the channel sits one more box: the noise source. Noise is anything that changes the signal between sending and receiving. On a phone line, that is static. In a lecture hall, it might be a cough, a bad microphone, or a word the audience does not know.

The model's strength is that it makes failure visible. If the message that arrives is not the message that was sent, you can ask where along the line it changed.

Its weakness is that it is one way. The receiver does nothing except receive. Wilbur Schramm, writing in 1954, redrew it as a loop: each person both sends and receives, and each interprets messages through their own field of experience. Later models add feedback and context.

So, two questions for us. First, is a one-way model still useful for anything, or has social media made it obsolete? Second, Weaver argued that the model could describe meaning, not just signals. Was he right to stretch an engineering diagram that far?

The scenes, one per paragraph:

1. A title card with the paper's name and 1948, and a small drawing of an old telephone.
2. Five boxes in a row, drawn left to right as each is named, joined by arrows.
3. The noise box dropping in above the channel, with a jagged arrow down into it.
4. The same row, with a question mark at the point where the message changed.
5. The straight line bending into a loop between two people, each with a thought bubble labelled field of experience.
6. The two questions, written large and left on screen.

That last scene is the important one. Leave it up while you talk. A seminar group discusses whatever is on the screen when the video stops.

Present it, or play it

There are two ways to use the video in the room, and it is worth deciding before you start.

Play it, then lead. The video runs for five minutes, stops on the questions, and you take over. This is the safest option if you are nervous, because the explanation is already finished and timed. Make sure the narration is loud enough for a room, not just for headphones.

Present over it. Mute the narration, or export the drawings and step through them, and speak the script yourself. Some tutors assess your spoken delivery, in which case this is the only option. The drafted script then becomes your speaking notes, already split into the same chunks as the scenes.

If the course is online, playing the video in a call works well, but share the file in the chat as well. Screen-shared video in a call often stutters, and people like to rewatch the model before answering.

Make it

1. 1

   Write the one sentence first

   State the idea in a sentence you could say in a corridor. If it takes two sentences, the topic is two presentations; pick one.

2. 2

   Write a paragraph per part and end on two questions

   One part of the model or one claim of the reading per paragraph. Close with two questions that have more than one defensible answer. Aim for 600 to 750 words for five minutes.

3. 3

   Draft it in openCanviz

   Paste the script, choose Keep my wording so the narration matches what you checked, set a five-minute target and pick whiteboard style, where each diagram builds as you name its parts.

4. 4

   Fix scenes that do not show their point

   Watch the draft once. Any scene where the drawing does not match its paragraph gets edited on its own; you do not regenerate the whole video.

5. 5

   Check names, dates and terms against the reading

   Author names, years, the exact terms the reading uses. Drafted text sometimes swaps a date or a name, and in a seminar the group has the source open.

6. 6

   Export and test the file offline

   Copy it to a memory stick or download it to the laptop you will use. Seminar room wifi is not something to depend on in the first minute.

What to cut when you are behind

If minute 20 arrives and the scenes are still rough, cut in this order:

• Cut examples before structure. A five-box diagram with one example beats three examples with no diagram.
• Cut the history before the critique. The group needs to know the model's weakness more than the name of every later theorist.
• Never cut the questions. A seminar presentation that ends on "any questions?" gets silence. One that ends on a specific question gets a discussion.
• Do not cut the check. Five minutes of checking names is worth more than five minutes of improving a drawing.

Free videos carry a small watermark. For a seminar warm-up that is usually fine. If the presentation is formally assessed and you want a clean file, a one-off payment unlocks a clean export of that single project.

Common questions

Is a video presentation allowed in a seminar? Usually, but ask. Some tutors want to assess how you speak, in which case present over the drawings yourself rather than playing the narration. If the brief says nothing, a short video followed by you leading the discussion is normally welcome.

How long should a seminar video presentation be? Five minutes for one model or one reading, unless you were told otherwise. That is 600 to 750 words of script. Longer eats into the discussion, which is the point of the seminar.

What if the reading is a long paper rather than a model? Pick the paper's main claim and the two or three moves that support it, one paragraph each. The guide to turning a research paper into a video summary covers how to choose which parts to keep.

Can I do this without slides at all? Yes. The drawn scenes do the job slides would, and they build in the order you speak, which slides rarely do. There is more on that in how to make a video presentation without PowerPoint.

What if this is a graded class project rather than a seminar? The priorities shift towards structure and sourcing. How to make a video for a school project covers what markers check.