Keep your script clear when changing the video direction

A script and a shot plan describe different parts of the same video. Mixing them together makes it harder to tell whether the presenter should say a line, perform an action or wait while text appears.

Keep the spoken words in one place and the visual instructions beside them. This also makes revisions easier to assess: changing the setting should not quietly change the product promise.

Label the words you want spoken

For a fictional app called Pinboard Desk, assume a user can save a link and add a note explaining why it matters. An original starting script could be:

“Found a useful link? Save it with a note about why it matters. When you return, the reason is right there beside it.”

That is dialogue. “Show a real recording of adding the note” is visual direction. A caption beneath the published post might invite viewers to see the full workflow. These are related pieces of copy, but they have different jobs.

When supplying your own dialogue in Personify Chat, label it clearly with “Use this script:” followed by the complete text. If you want a wording change, describe that change separately. If you want only a visual change, say which words should stay.

Change one part of the brief deliberately

A useful follow-up for the fictional example could be: “Keep that script word for word. Start with the saved-link screen, then show the note being added. Use a real recording for those steps.”

That instruction changes the visual sequence while keeping the explanation. It does not grant permission to invent interface controls or replace the note feature with a broader claim about organizing a whole business.

If you do want a shorter script, specify what must survive. In this example, the note’s purpose matters more than a general claim about productivity. Compare the revision with that requirement before accepting it.

Decide whether the words are spoken or shown

A text-led reel can have words that appear over silent activity. A talking-head video has a presenter delivering dialogue. Neither is automatically the right choice for every idea.

Before approving the visual plan, confirm which experience you are choosing. If you expect speech, the brief should contain complete sentences the presenter can say naturally. If you expect silent action, the screen text should be short enough to read while following the image.

A post caption is separate again. Do not assume text written for the social post will be spoken in the video or placed on its frames.

Check the saved brief, then the output

Open the brief and inspect the title, complete script, selected avatar, format and product references together. Follow the duration and word guidance for that format. If the full explanation does not fit, shorten it deliberately or split the story into separate videos.

Reviewing a script is different from approving a generation quote. Check both before creating a take.

After generation, listen to the actual speech and inspect the visual sequence. Correct source text is necessary, but it does not guarantee that every generated word or action is correct. A precise brief gives you something concrete to compare with the finished result.