Explain through images

How can you turn a procedure into a short, understandable video that is ready for production?

1250 € per session, excl. VAT

In short

In a half-day session of 3 hours you use generative AI to turn a procedure into a visual scenario. You prepare a timed script, a storyboard and production instructions, then check comprehension, accessibility and the associated rights before any production or distribution.

Why this training?

An explainer video can be appealing without enabling the viewer to take action. Defining what the viewer needs to understand, remember and do avoids overly long introductions, unnecessary details and purely decorative images. AI helps simplify text and suggest shot breakdowns, but it can invent a step or suggest a misleading representation. The task is to check each proposal against the source procedure and distribution constraints. Running case study: The fictitious organisation « Avelune Réemploi » lends DIY equipment. It wants to explain to new members how to return an item of equipment: check the accessories, report a fault and leave the equipment at the counter. A fictitious procedure serves as the reference for preparing a video lasting 60 to 90 seconds. Application at work: Within the following two weeks, take an anonymised procedure from your organisation, adapt the design brief and ask a colleague who is representative of the audience to explain the steps in their own words before starting production.

Objectives of this training

  • Define the audience, the central message and the expected action.
  • Use AI to write a script that is faithful to a source procedure.
  • Break the story down into shots combining narration, images and on-screen text.
  • Identify the rights, consents and notices that need to be checked.
  • Test comprehension and correct any ambiguities identified.

Who is it for and what are the prerequisites?

Roles and tasks concerned

  • Communications officers producing short educational content.
  • Staff who need to explain a service, a way of using something or a procedure through video.

Prior skills

  • You have already formulated and refined a request to an AI assistant.
  • You can distinguish the essential steps in a simple procedure; no editing skills are required.

What you will work on

  1. Define the scope of the explanation

    Objective: Choose what the viewer should be able to do. Novice audience, viewing context, a single message, expected action and essential information.

  2. Write to be heard

    Objective: Produce concise and accurate narration. Situation–steps–recap structure, sentences suited to speech, accessible vocabulary and duration estimated by reading aloud.

  3. Move from text to shots

    Objective: Make each step visible without overload. Framing, continuity, synchronisation, subtitles and the choice between diagrams, fictitious screenshots or filmed footage.

  4. Test before production

    Objective: Identify misunderstandings and ensure a safe transition to production. Reviewing the storyboard, restating it without assistance, checking readability and ensuring corrections are traceable.

What you take away

  • Scoping sheet: audience, message, duration and distribution.
  • Timed script with narration and on-screen text.
  • Storyboard of six to eight shots and production instructions.
  • Completed checklist and comprehension test report.

Method and assessment

  • Short teaching inputs followed by immediate practice using the running case study.
  • Individual work followed by work in pairs, with participants reviewing each other’s outputs.
  • Guided group discussion: what works, what needs correcting and what remains a human decision.
  • Preparation: Arrange access to a tool authorised by your organisation, with its terms of use checked. Bring a fictitious or anonymised procedure and specify the intended distribution channel, without personal data or confidential documents.

Observable criteria: The three return steps are present, in order, with no unjustified additions. ; The timed reading of the script meets the target of 60 to 90 seconds. ; Each shot supports an identifiable piece of information, with readable text and planned subtitles. ; Each proposed resource has a documented status: usable, requiring permission or to be replaced. ; A peer restates the steps without assistance; any errors identified lead to a documented correction..

Frequently asked questions

Will you leave with an edited film?

No, the deliverable is an operational design brief. Filming, media generation and editing take place afterwards.

Do you need to generate images during the session?

No. Sketches and shot descriptions are sufficient to validate the visual logic.

Can you reproduce a voice or a face?

Not in the exercises. Any real-world use requires appropriate rights and consents, among other things; no impersonation or misleading deepfakes.

How can you assess quality without a finished video?

We measure fidelity to the source text, spoken duration and comprehension of the storyboard. The audiovisual output will need to be checked after production.

Last updated: