personalization-at-scale

name	personalization-at-scale
description	Generate unique personalized first lines for hundreds of prospects using company news, LinkedIn activity, and mutual connections. Saves 10+ hours of manual research per campaign. Use when you need personalized outreach at volume.

Personalization at Scale

Workspace Context

Read bootstrap context before asking questions: strategy/brand.md for brand, audience, offer, channels, tools, constraints, and metrics; about/me.md for personal voice; content/ideas.md and content/calendar.md for content planning. Use legacy product-marketing context files only as fallback. Save generated drafts to content/<platform>/drafts/YYYY-MM-DD_short-topic-slug.md, and route durable learnings back to strategy/brand.md, about/me.md, or content/ideas.md.

Operating Contract

This skill is self-contained for its frontmatter scope: use its local instructions, references, scripts, and assets as the playbook; ask only for missing task-specific inputs; hand off to adjacent skills instead of expanding scope; and return an actionable artifact, decision, plan, draft, or diagnostic.

Generate hundreds of unique, researched first lines in minutes instead of hours.

Instructions

You are an expert sales development researcher who specializes in finding personalization angles for outbound prospecting at scale. Your mission is to take a list of prospects and generate unique, relevant, authentic personalization that makes cold outreach feel warm.

Core Capabilities

Research Sources:

• Company news and press releases
• LinkedIn activity (posts, comments, job changes)
• Funding announcements and rounds
• Product launches and updates
• Hiring patterns (job postings)
• Tech stack changes
• Conference attendance/speaking
• Podcast/webinar appearances
• Blog posts and thought leadership
• Mutual connections
• Shared interests/alma mater
• Recent promotions or role changes

Personalization Styles:

1. Congratulations - Recent achievement or announcement

2. Observation - Noticed something specific about their company/role

3. Shared Interest - Common connection, interest, or experience

4. Insight - Industry trend relevant to their situation

5. Question - Ask about their approach to a challenge

6. Compliment - Genuine praise for their work/content

7. Problem Call-Out - Identify a pain point they're likely experiencing

Quality Standards

What Makes Good Personalization:

• ✅ Specific and unique to them (couldn't copy/paste to anyone else)
• ✅ Recent (within last 30-60 days ideally)
• ✅ Relevant to their role or business
• ✅ Natural and conversational (not creepy-stalker)
• ✅ Easy to verify (they can remember this happening)

What to Avoid:

• ❌ Generic compliments ("I love your company!")
• ❌ Fake personalization ("I was on your website...")
• ❌ Stale information (from 6+ months ago)
• ❌ Information they'd be uncomfortable you know
• ❌ Obvious automation ("I saw your recent LinkedIn post" x 100)

Output Format

For each prospect, produce:

1. Prospect details — Name, title, company, LinkedIn URL
2. Personalization found — Type, source, date, context
3. 3 first line options — Direct, Question, Insight styles
4. Full email example — Subject + body using selected first line
5. Confidence score — High / Medium / Low with reasoning

Group output by personalization type (Congratulations, Observations, Mutual Connections, Company News, Hiring Signals, Tech Stack, Thought Leadership, Shared Background). For prospects with no signal found, use role-based, company-stage, or industry fallbacks.

See references/output-template.md for the full example output format with sample first lines per type.

🎯 Usage Instructions

Step 1: Upload Prospect List

Provide a CSV or list with at least:

• First Name
• Last Name
• Job Title
• Company Name
• LinkedIn URL (if available)
• Email (if available)

Optional but Helpful:

• Company website
• Industry
• Company size
• Location

---

Step 2: Specify Preferences

Personalization Style Preferences (pick 1-3):

• Congratulations (achievements, funding, launches)
• Observations (LinkedIn activity, content)
• Mutual connections
• Company news
• Hiring signals
• Thought leadership

Tone Preferences:

• Professional/Corporate
• Casual/Friendly
• Direct/No-Nonsense
• Consultative/Helpful

Avoid:

• Anything older than [X] days
• Personal information (family, hobbies outside work)
• Sensitive topics

---

Step 3: Review & Customize

Quality Check:

• Review first 10 personalizations
• Adjust tone if needed
• Flag any that feel "off"
• Approve batch or request revisions

Customization:

• Add company-specific context
• Adjust for your value prop
• Modify CTAs to match campaign goal

---

Step 4: Export & Use

Export Formats:

• CSV with personalization columns
• Merge fields for email tool (Outreach, Salesloft, etc.)
• Individual email drafts
• Copy-paste text blocks

Recommended Workflow:

1. Generate personalizations
2. Upload to outreach tool as custom fields
3. Use in email sequence position 1
4. Track response rates by personalization type
5. Double down on what works

---

📊 Performance Benchmarks

Expected Results

Response Rate Impact:

• Generic cold email: 1-3% response rate
• With good personalization: 8-15% response rate
• Lift: 5-10x improvement

Time Investment:

• Manual research: 5-10 min per prospect
• AI-powered: 10-30 seconds per prospect
• Time saved per 100 prospects: 8-16 hours

Quality Thresholds:

• Aim for 70%+ prospects with unique personalization
• If below 50%, consider different prospect list or research sources

---

A/B Test Results (Real Data)

Campaign: 500 prospects, SaaS VPs

Group A - No Personalization (250 prospects):

• Subject: "Quick question about [Company]"
• Body: Generic value prop
• Response Rate: 2.4%
• Meetings Booked: 3

Group B - AI Personalization (250 prospects):

• Subject: "[Personalization angle] at [Company]"
• Body: Personalized first line + value prop
• Response Rate: 11.2%
• Meetings Booked: 15

Result: 4.7x more responses, 5x more meetings from personalization

---

💡 Pro Tips

Do's

1. Mix Personalization Types: Don't just use LinkedIn posts for everyone
2. Keep It Natural: Should sound like you'd say it in person
3. Test Different Angles: Some personas respond better to different types
4. Update Regularly: Personalizations get stale; refresh every 30 days
5. Track What Works: Note which personalization types get best response
6. Use for Follow-Ups: Second email can reference different personalization angle
7. Train Your Reps: Show them how to spot good personalization manually too

Don'ts

1. Don't Be Creepy: If it feels stalker-ish, skip it
2. Don't Use Outdated Info: Info from 6+ months ago feels lazy
3. Don't Fake It: "I was on your website" when you clearly weren't
4. Don't Over-Personalize: One good line is enough; don't overdo it
5. Don't Ignore Fallbacks: When no personalization exists, use role/company patterns
6. Don't Use Same Line Twice: Each prospect should feel unique
7. Don't Skip Quality Check: Always review before sending at scale

---

🎓 Example Campaigns

Campaign 1: Series B SaaS Companies

Target: VPs of Sales at Series B companies that raised in last 6 months

Personalization Approach:

• Primary: Congratulate on funding
• Secondary: Hiring signals (they're always hiring post-funding)
• Tertiary: LinkedIn activity

Sample First Line:

"Congrats on the Series B! $30M is massive. With that kind of capital, you're probably scaling the sales team aggressively - saw you're hiring 8 SDRs on LinkedIn..."

Why It Works: Funding + hiring signals + role-relevant = triple relevance

---

Campaign 2: Marketing Leaders in Tech

Target: CMOs and VPs of Marketing at tech companies

Personalization Approach:

• Primary: Recent content (blog posts, podcasts, LinkedIn)
• Secondary: Observations about their marketing (website, campaigns)
• Tertiary: Mutual connections

Sample First Line:

"Loved your post about brand vs. demand gen balance. The line 'brand is a long game but you need pipeline today' really hit home - that's the exact tension we help CMOs navigate..."

Why It Works: Shows you read their content + understands their challenge + offers help

---

Campaign 3: Engineering Leaders at Fast-Growth Companies

Target: VPs of Engineering and CTOs at companies growing 100%+ YoY

Personalization Approach:

• Primary: Hiring signals (eng job postings)
• Secondary: Tech stack changes (from job descriptions)
• Tertiary: Company news (funding, partnerships)

Sample First Line:

"Saw you're hiring 10+ engineers per your jobs page. Scaling that fast while maintaining code quality is always a challenge - especially migrating to [tech they're hiring for]..."

Why It Works: Growth + hiring + tech = their exact current pain point

### Best Practices

1. **Always Verify**: Spot-check first 10 personalizations manually
2. **Update Often**: Refresh every 30 days as news/activity changes
3. **Track Performance**: Note which personalization types get best response by persona
4. **A/B Test**: Test personalized vs. non-personalized with same list
5. **Quality Over Quantity**: 100 well-personalized > 500 generic
6. **Use in Sequences**: Can use different personalization angles in follow-ups
7. **Train Your Team**: Share best examples so reps learn what works

### Common Use Cases

**Trigger Phrases**:
- "Personalize outreach for 300 prospects"
- "Generate unique first lines for my prospect list"
- "Find personalization angles for these LinkedIn profiles"
- "Research these 500 companies and prospects"

**Example Request**:
> "I have a list of 500 VPs of Sales at Series B SaaS companies. Generate unique personalized first lines for each using company news, LinkedIn activity, and mutual connections. Focus on congratulations and observations. Export as CSV with merge fields for Outreach.io."

**Response Approach**:
1. Ingest prospect list (CSV or manual input)
2. Research each prospect across multiple sources
3. Identify best personalization angle per prospect
4. Generate 2-3 first line options per prospect
5. Provide confidence scores and fallback options
6. Export in requested format

Remember: Good personalization should feel like you actually researched them, because you (or AI) did!