Training video scripts

What a training video script is

A training video script is the spoken text and on-screen detail for an instructional video, written so that one video teaches one task and a person can do that task afterwards without asking anybody. It carries a stated prerequisite at the front, chapter markers through the middle, and a check at the end the viewer can measure themselves against.

It is not a recorded presentation. It is also not a demo script, which is written to persuade somebody who has not bought yet; a training script assumes the decision is made and the viewer has a job to do this morning. And it is not the written manual. That is training materials, and most businesses need both, because some people will not watch a video and some will not read a page.

When you need one

You need training video scripts when the same explanation is being given over and over by the one person who understands it. New staff, a new system, a seasonal intake, a process only the owner can do properly. Video suits anything where watching somebody do it is faster than reading about it.

The second case is a business that already recorded training and found nobody finishes it. That is nearly always a structure problem, not a production problem. One long unstructured recording cannot be searched, so nobody returns to it.

You do not need video at all if the task is a checklist. A one-page procedure is faster to write, faster to change and faster to follow than four minutes of somebody talking. Write it as an SOP and film only the parts where seeing it matters. I will say so on the first call.

What goes in it

The design assumption is that nobody watches training video the way it was recorded. It gets watched at one and a half times speed by somebody who has already done the task twice, paused while they try the step themselves, rewound ten seconds because a field name went by too fast, and reopened three weeks later for one answer. Everything below exists to survive that.

• One task per video. Three to six minutes, covering a single thing a person can now do. A twenty-minute video covering seven tasks is seven videos that have not been separated yet.
• A stated prerequisite. The first line names what the viewer must already have or know: the login, the completed form, the previous video. It saves four wasted minutes watching something you cannot follow.
• Chapter markers. Each step marked with its time, listed at the top and on the page beneath the video, so a person returning for one answer finds it in seconds.
• Field names said exactly. Buttons, menus and fields spoken word for word as they appear, and shown as on-screen text at the same moment, because that is what gets missed at speed.
• A check at the end. One sentence describing what the screen or the bench should look like if it worked, so the viewer knows whether to move on or watch again.

What you get

The scripts. One two-column document per video, formatted for reading aloud: spoken text on the left with a running time at every step, screen or shot notes on the right. Each is read aloud against a stopwatch before delivery. Writing for the ear is not writing for the page, and a procedure document read aloud sounds wrong immediately, because it is built for a reader who can go back over a clause.

The chapter list. Timings and titles for each video, formatted to paste straight into the player description, whether you host internally or on YouTube.

The sequence. A one-page running order for the whole set, showing which video depends on which, so somebody joining the company knows what to watch and in what order.

A template. After the first set, the structure as a blank document, so that when your own people record the next procedure the videos still belong together.

A worked example

An illustration, not a client. A Winnipeg food distributor with about forty staff puts in a new handheld scanner system. The supervisor records one training video. It runs twenty-six minutes, covers receiving, picking, returns and stock counts, and begins with four minutes of context. Three months later new staff are still being walked through the floor by hand.

The same material is rewritten as seven scripts of three to five minutes. Receiving becomes two videos, because a scheduled delivery and an unexpected one are different tasks with different screens. Each opens with the prerequisite, which for five of them is that the handheld is logged in and paired. Each ends with the check: the pallet count matches the manifest and the light on the unit is green. The stock count video gets five chapter markers, because that is the one people return to. The set is watched in order once by a new hire, and the picking video alone is reopened constantly, which is the point.

How it runs

1. A half-hour call. What has to be taught, who to, how they currently learn it, and whether any of it is already written down. No charge.
2. A fixed price in writing. Number of videos, run times and whether a written counterpart is included, agreed before anything starts.
3. Watching it done. I need to see the task performed once, on a call or a phone recording, and to try it myself where possible. Scripts written from a manual describe the manual, not the job.
4. The task list. Before any writing, you approve the split: which tasks get their own video, in what order, with what prerequisites.
5. The drafts. Roughly two working days per video once the split is agreed, delivered in batches rather than all at the end.
6. A test. Somebody who does not know the task follows a script while it is read to them. Whatever they get stuck on is rewritten.

What it costs

Training video scripts are quoted as a fixed price in CAD after the half-hour call, usually per video within an agreed set, so you can see what adding or dropping one changes. The price depends on how many tasks there are, how much of the process already exists in writing, and whether the written version is produced alongside. The pricing page explains how fixed pricing works. The figure is agreed in writing before anything starts.

What happens next

Three things usually follow. The set gets a written counterpart, because some staff will never watch a video, and it is drawn from the same scripts so the two cannot contradict each other. Staff who cannot use sound on the floor need proper subtitle files rather than machine output, since the field names are exactly what machines get wrong. And once the videos exist, the process behind them is often worth writing down properly as procedures, so it outlives whoever invented it. None of that is assumed. The scripts are a complete piece of work on their own, and you can record them and stop.