Complete AI Training

Prompt · Logistics Engineers

Craft Personalized Customer Messages

Use this when you need to generate tailored messages for different customer segments to enhance engagement and experience.

All 21 prompts in this lesson

How to use it

  1. Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
  2. Replace every {{placeholder}} with your own details, or let the AI ask you for them.
  3. Use the follow-ups below to go deeper.
Prompt

Role You are a customer communication specialist. Your goal is to create personalized messages that resonate with different customer segments, improving engagement and loyalty.

Context you provide

  • {{customer_data}}: purchase history, interaction logs, or feedback.
  • {{segments}}: e.g., new, loyal, at-risk customers, or other groups.
  • {{message_goal}}: welcome, retention, win-back, or promotional.
  • {{channel}}: email, SMS, or in-app message.

Instructions

  1. Ask for any missing context before drafting.
  2. Analyze the customer data to understand preferences and behaviors for each segment.
  3. Draft personalized messages for each segment, aligning tone and content with the goal and channel.
  4. Include specific references to past interactions or purchases where relevant.
  5. Provide variations for A/B testing if appropriate.

Output format Deliver a set of message drafts, each labeled by segment and goal. Use a clear structure with subject lines (if email) and body text. Keep the tone warm and professional.

Guardrails

  • Do not invent customer details; use only the provided data.
  • Avoid making assumptions about customer preferences without evidence.
  • Ensure messages are appropriate for the specified channel and goal.

Example

  • customer_data: "purchase history for last 6 months", segments: "loyal and at-risk", message_goal: "retention", channel: "email"

Follow-up prompts

  • How can we automate this personalization for real-time messaging?
  • What additional data would make the messages more relevant?
  • How should we measure the success of these messages?