How to Turn One Testimonial Video Into 3 Vertical Ad Variations With AI

The problem and who this is for

One testimonial can support several ad variations, but most teams only cut one generic version and stop there. This workflow is for agencies, in-house growth teams, creators making ads, and editors repurposing customer proof into vertical video. The goal is to get to three short ad variations that keep the real proof but change the hook, framing, and call to action with the fewest steps possible.

Prerequisites

• Access to ChatGPT
• A source file or capture workflow that matches the article
• A place to save the final output, such as a Google Doc, notes app, spreadsheet, or editor brief
• An editing or publishing workflow for your final short-form asset

How to capture or gather the source material

1. Start with the testimonial video transcript or a clean written summary of the customer's actual words.
2. Add one short product or offer brief that explains the audience, the core problem, and any compliance or brand limits.
3. Remove private customer details and anything you cannot substantiate in paid or branded creative.
4. Mark the strongest proof lines in the transcript. Look for clear before-and-after language, objection-handling moments, or concrete outcomes.

Step-by-step workflow

1. Upload the testimonial transcript and offer brief into ChatGPT.
2. Ask for three ad-framing options built from the same proof source. Good variation levers include different opening hooks, different audience entry points, or different emotional emphasis.
3. Tell ChatGPT to keep the original proof intact and vary the framing around it. You want three distinct ads, not three paraphrases of the same script.
4. Have it draft three vertical ad scripts with hook, body, proof line, and call to action.
5. Check each script against the original testimonial and the brief. The strongest version is often the one that preserves the customer's own words most clearly.

Tool-specific instructions

Primary tool: ChatGPT

• ChatGPT is a strong primary choice because the main job is variation generation from a limited but meaningful source set.
• Ground the variation on framing, not invented proof. The source testimonial should still do the heavy lifting.
• Keep the three versions meaningfully different. One can open with the problem, one with the result, and one with the objection or skeptic angle.

Alternative: Claude

• Claude is a good alternative when you want a cleaner editorial pass and less ad-sounding copy from the same source inputs.
• It is especially good for making sure the proof remains believable and restrained.

Alternative: Gemini

• Gemini is useful if your source set includes Drive documents, screenshots, or quick visual references you want to review alongside the transcript.
• It can also be handy if part of the source material starts on mobile.

Copy and paste prompts

ChatGPT testimonial variation prompt

{
  "role": "You are turning one testimonial source into three distinct vertical ad directions.",
  "goal": "Create three ad framings from the same proof source without inventing new claims.",
  "rules": [
    "Use only the proof points in the source material.",
    "Vary the hook, framing, and call to action.",
    "Keep the customer meaning intact."
  ],
  "output_fields": [
    "variation_name",
    "core_hook",
    "proof_line_to_anchor_on",
    "audience_entry_point",
    "call_to_action_style"
  ]
}

ChatGPT testimonial ad script prompt

{
  "role": "You are writing vertical ad scripts from approved testimonial-based directions.",
  "goal": "Write three short ad scripts that preserve the original proof while changing the framing.",
  "requirements": [
    "Keep the language believable.",
    "Do not invent outcomes or customer details.",
    "Use spoken lines that can work in a vertical ad format."
  ],
  "output_fields": [
    "script_title",
    "opening_hook",
    "body_lines",
    "proof_line",
    "call_to_action",
    "editor_note"
  ]
}

Quality checks

• Each variation uses the same proof source honestly.
• The three openings are meaningfully different.
• No version overstates the testimonial or adds unsupported claims.
• The strongest customer language has been preserved where possible.

Common failure modes and fixes

• The three ads feel too similar: Force a different audience entry point or emotional frame for each one.
• The proof line gets rewritten until it sounds fake: Bring the original customer wording back in.
• The scripts make claims beyond the source: Cut anything not supported by the transcript or brief.
• The testimonial itself is weak: Do not force three ads from thin proof. Pair it with stronger customer language or a narrower angle.

Sources Checked

• OpenAI Help: File Uploads FAQ: https://help.openai.com/en/articles/8555545-file-uploads-faq (accessed 2026-03-25)
• OpenAI Help: ChatGPT Image Inputs FAQ: https://help.openai.com/en/articles/8400551-chatgpt-image-inputs-faq (accessed 2026-03-25)
• Claude Help: Uploading files to Claude: https://support.claude.com/en/articles/8241126-uploading-files-to-claude (accessed 2026-03-25)
• Claude Help: What are projects?: https://support.claude.com/en/articles/9517075-what-are-projects (accessed 2026-03-25)
• Gemini Help: Upload and analyze files in Gemini Apps on Android: https://support.google.com/gemini/answer/14903178?co=GENIE.Platform%3DAndroid&hl=en (accessed 2026-03-25)
• Gemini Help: Upload and analyse files in Gemini Apps on Computer: https://support.google.com/gemini/answer/14903178?hl=en-NA&visit_id=639100585389477783-308453270&p=code_upload&rd=1 (accessed 2026-03-25)

Quarterly Refresh Flag

Review by 2026-06-23. Re-check the current tool interface, upload behavior, supported source types, and any changes that affect this workflow before republishing or refreshing the article.