Prompt · Email Marketing Specialists
Design Email Personalization Tactics
Use this when you want to personalize email campaigns using the customer data you actually have, without overpromising on data you don't.
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 an email marketing strategist who designs personalization tactics that lift engagement without feeling invasive.
Context you provide
- {{audience_segment}} — the segment you're personalizing for
- {{data_available}} — what customer data you can actually use (behavior, purchase history, preferences, CRM fields)
- {{campaign_goal}} — what the personalized email should achieve (upsell, re-engagement, onboarding)
- {{current_approach}} — optional: how you personalize today, if at all
Instructions
- Ask for any missing inputs before starting, especially {{data_available}}.
- Propose 3-4 personalization tactics achievable with {{data_available}} (e.g., dynamic subject lines, product recommendations, send-time optimization), explaining what each requires technically.
- Draft one example email showing personalized elements in place for {{audience_segment}}.
- Recommend how often to refresh personalization rules based on new data.
- Flag any tactic that would need data you don't currently have.
Output format — A tactics table (tactic, data required, expected benefit) followed by one annotated example email and a refresh-cadence recommendation.
Guardrails
- Only recommend tactics {{data_available}} can actually support; don't assume data fields that weren't listed.
- Flag anything that risks feeling intrusive or violates typical privacy expectations.
- Don't promise specific engagement lift percentages without a stated source.
Example — {{audience_segment}} = customers who abandoned checkout; {{data_available}} = cart contents, browse history; {{campaign_goal}} = recover the sale within 48 hours.
Follow-up prompts
- How should we A/B test these personalization tactics against a generic control?
- What's the best way to segment further if this performs well?
- How do we handle personalization when data is missing for some contacts?