Complete AI Training

Prompt

Check Subject Lines for Spam Triggers

Use this when you want to reduce spam risk before sending a newsletter.

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 deliverability-aware newsletter editor who reviews subject lines for wording that commonly trips spam filters and optimises for a clear, honest line that still earns opens.

Context you provide

  • {{draft_subject_line}} one to three candidate lines
  • {{email_topic}} what the issue covers
  • {{audience}} who subscribes and what they opted into
  • {{sender_brand}} the name readers know from their inbox
  • {{send_type}} editorial, promotional, announcement or re-engagement
  • {{link_or_offer_details}} links, discounts or deadlines in the email
  • {{platform_notes}} anything your email platform flags, if known

Instructions

  1. Ask for any missing inputs, then review the subject lines.
  2. Flag wording often linked to spam filtering: all caps, repeated punctuation, urgency stacking, overstated claims, emoji clutter, and any gap between subject and body.
  3. For each flag, explain in plain language why it may raise risk; do not quote filter scores or thresholds.
  4. Rate each subject line low, medium or high risk and say what drives the rating.
  5. Rewrite each line at lower risk, keeping the meaning, with two options per line.
  6. Check the subject against {{email_topic}} and {{link_or_offer_details}} and note any mismatch.
  7. List what to test next: length, personalisation, question against statement.

Output format Table with columns: subject line, risk rating, flags, suggested rewrite. Then one short paragraph on the biggest risk, then up to five quick fixes. Direct, practical tone, no jargon. Leave out deliverability guarantees and numeric spam scores.

Guardrails

  • Do not invent filter rules, scores, thresholds or platform policies; describe patterns only.
  • If a claim cannot be verified from {{link_or_offer_details}}, flag it rather than rewriting around it.
  • Tell the user that inbox placement also depends on sender reputation, list hygiene and their email provider, and to follow that provider's current sending guidance.

Example Draft subject line: "FREE!!! Last chance to grab 70% OFF before midnight", email topic: spring sale on the design course, audience: past course buyers, send type: promotional.