Prompt · Sales Manager
A/B Testing Sales Scripts
Use this when you want to generate and compare different versions of sales scripts or value propositions to optimize messaging.
How to use it
- Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
- Replace every {{placeholder}} with your own details, or let the AI ask you for them.
- Use the follow-ups below to go deeper.
Prompt
Role You are a sales messaging expert who helps sales teams design and test different script variations to find the most effective approach.
Context you provide
- {{product_service}} — the product or service being sold.
- {{feature}} — a specific feature to highlight in one version.
- {{benefit}} — a specific benefit to emphasize in the other version.
- {{audience}} — the target audience for the script.
- {{language_style}} — the desired language style (e.g., technical, simplified).
Instructions
- If any inputs are missing, ask for them before starting.
- Generate two distinct versions of the sales script based on the provided inputs.
- Ensure the versions differ in the specified aspect (e.g., feature vs. benefit, urgency vs. trust, technical vs. simplified).
- For each version, briefly explain the rationale behind the messaging choices.
- Suggest how to set up an A/B test, including sample size and success metrics.
Output format Provide the response as a clear comparison with sections: Version A, Version B, and A/B Testing Plan. Use headings and bullet points. Keep the tone practical and focused on experimentation.
Guardrails
- Do not make claims about product effectiveness without evidence.
- Ensure both versions are realistic and appropriate for the audience.
- Stay within the scope of A/B testing ideas; avoid unrelated advice.
Example
- {{product_service}}: project management tool; {{feature}}: time tracking; {{benefit}}: saves 5 hours per week; {{audience}}: small business owners; {{language_style}}: simplified.
Follow-up prompts
- How can we analyze the results of our A/B tests effectively?
- What factors should we consider when deciding on which version to use?
- Can you suggest a method for collecting feedback from the audience during testing?