You get:
- lazy personalization (just name and company)
- no genuine insight (doesn’t show research)
- personalization that feels templated
- missed opportunities to connect
- prospects who feel like a number
But real personalization is not a token.
It is evidence you did your homework.
- LinkedIn activity: posts, comments, job changes
- Company news: funding, hires, product launches
- Mutual connections: people you both know
- Shared interests: groups, topics, events
- Recent achievements: awards, recognition, milestones
Without real personalization, you’re just spam.
This framework forces AI to find authentic personalization hooks.
Assume the role of a cold email researcher who finds authentic personalization hooks. Your task is to identify personalization opportunities. Generate: 1. LINKEDIN INSIGHTS - Recent posts or articles - Job history highlights - Shared connections - Groups or interests 2. COMPANY INSIGHTS - Recent news (funding, hires, launches) - Company milestones - Current challenges (inferred) 3. MUTUAL CONNECTIONS - People you both know - How to reference them 4. SHARED INTERESTS - Industry topics they engage with - Events they attended - Content they shared 5. PERSONALIZATION HOOKS (5-7) - Specific opening lines referencing research - Why each would resonate 6. PERSONALIZATION RECOMMENDATIONS - Best hook to use - What to avoid INPUTS: Prospect Name: [INSERT] Prospect Role: [INSERT] Prospect Company: [INSERT] Prospect LinkedIn URL (or describe profile): [INSERT OR "UNKNOWN"] Your Industry: [INSERT] Your Mutual Connections (if any): [LIST OR "NONE"] Recent Company News (if known): [LIST OR "UNKNOWN"] RULES: - Find specific details (not generic observations) - Reference recent activity (last 30-90 days) - Avoid creepy or stalker-level detail - Personalization must be authentic (not "I see you're in marketing") - One strong personalization hook is better than three weak ones - If no personalization available, focus on problem/outcome
- Spend 2-3 minutes researching each prospect before writing.
- Find something specific from the last 30 days.
- One genuine insight > three generic mentions.
- Avoid being creepy (don’t mention their kids’ names or home address).
- If no personalization available, focus on the problem they likely have.
Prospect Name: Sarah Chen
Prospect Role: VP of Sales
Prospect Company: ScaleFlow
Prospect LinkedIn URL: /in/sarahchen — recent post about hiring sales reps, shared an article about sales automation
Your Industry: Sales tech (CRM automation)
Your Mutual Connections: Mark from Acme (sales leader)
Recent Company News: ScaleFlow just announced 50% growth, hiring 10 sales reps
This framework improves outcomes by forcing:
- LinkedIn research (recent activity)
- company insight gathering (timely news)
- mutual connection identification (trust signals)
- shared interest discovery (common ground)
- authentic hook development (credibility)
Great cold emails don’t just use your name — they show they did their homework.
Build Better AI Systems
Subscribe for advanced prompt engineering, AI sales tools, cold email frameworks, and practical strategies for sales professionals and founders.