Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Generate cold email templates, LinkedIn messages, follow-up sequences, and value propositions tailored to specific industries that actually get responses and book discovery calls.
.claude/skills/nicepkg-cold-outreach-generator/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -7% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 6% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -6% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 67% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 63% | 0% |
Get responses. Book calls. Win clients.
Input: Target industry + your solution Output: Cold email templates, LinkedIn messages, follow-up sequences
Subject: Specific Pain Point Question] Example: "Managing 200+ inspections/month in Excel?"
Body (5 sentences max):
Sentence 1: Specific observation about their business "I noticed Company] handles fire inspections for commercial buildings in City]."
Sentence 2: Common pain point "Most inspection companies I work with struggle with specific problem]."
Sentence 3: Brief solution hint "I build custom software that specific benefit]."
Sentence 4: Social proof "Built similar system for Company] - saved them 15 hours/week."
Sentence 5: Soft CTA "Worth a 15-min call to see if this makes sense for you?"
Signature Your Name] Your Title] Calendar Link]
"Hi Name], I work with industry] companies to solve specific problem]. Thought it might be worth connecting."
Follow-up after connection: "Thanks for connecting! I noticed Company] does their work]. Quick question - how do you currently handle their pain point]? Most industry] businesses I talk to struggle with that."
Day 0: Initial outreach Day 3: "Hi Name], following up on my note from Monday. Curious if pain point] is something you're dealing with?" Day 7: "Last note from me! If problem] isn't a priority right now, no worries. But if it is, happy to show you what we built for Similar Company]."
Stop after 3 touches. Don't be annoying.
✅ Specific: "Managing HVAC service calls" not "running your business" ✅ Relevant: Mention their actual work ✅ Brief: 5 sentences max ✅ Social proof: Name similar companies you helped ✅ Easy CTA: Calendar link, not "let me know when you're free"
❌ Generic: "I help businesses save time and money" ❌ Long: 3+ paragraphs ❌ Salesy: "Limited time offer! Act now!" ❌ Desperate: "Please respond!" ❌ Hard pitch: Pricing and features in first email
Template #1: The Observation Subject: Saw you're hiring role]
Body: "Hi Name], noticed you're hiring a dispatcher/admin/etc].
That's usually a sign you're growing fast and current systems can't keep up.
I build software for industry] that lets you grow without hiring more admin staff. Built similar system for Company] - they handle 2x the volume with same team now.
Worth a quick call?"
Template #2: The Problem Subject: Pain point question]
Body: "Hi Name],
Quick question: How do you currently painful process]?
Most industry] companies waste 10-15 hours/week on this manually.
I build custom software that automates it. Company] used to spend 15 hours/week, now it's automated.
15-min call to show you how it works?"
Template #3: The Referral Subject: Mutual connection] suggested I reach out
Body: "Hi Name],
Mutual Connection] mentioned you might be interested in solution to pain point].
I recently built specific software] for their competitor]. Saved them 20 hours/week and $80K/year.
Worth exploring if similar solution makes sense for Their Company]?"
Outreach Math:
To get 1 client/month: Send 20-25 outreach emails/week
Cold outreach is a numbers game. Most won't respond. That's okay.
Focus on:
Do this every week. You'll book calls. Calls turn into clients.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 22,566 | 12,867 | -43% | 1 | 1 | 0% | 3,226 | 3,006 | -7% | 0 | 0 | — |
case-02 | fail→pass | 18,129 | 11,798 | -35% | 1 | 1 | 0% | 2,795 | 2,970 | +6% | 0 | 0 | — |
case-03 | fail→pass | 19,347 | 10,416 | -46% | 1 | 1 | 0% | 2,999 | 2,818 | -6% | 0 | 0 | — |
case-04 | pass→pass | 19,125 | 16,267 | -15% | 1 | 1 | 0% | 2,801 | 3,265 | +17% | 0 | 0 | — |
case-05 | pass→pass | 16,503 | 12,073 | -27% | 1 | 1 | 0% | 2,554 | 2,959 | +16% | 0 | 0 | — |
case-06 | pass→pass | 19,081 | 16,737 | -12% | 1 | 1 | 0% | 3,093 | 3,723 | +20% | 0 | 0 | — |
case-07 | fail→pass | 11,523 | 12,458 | +8% | 1 | 1 | 0% | 1,810 | 3,024 | +67% | 0 | 0 | — |
case-08 | fail→fail | 18,001 | 13,904 | -23% | 1 | 1 | 0% | 2,834 | 3,236 | +14% | 0 | 0 | — |
case-09 | fail→fail | 10,910 | 7,141 | -35% | 1 | 1 | 0% | 1,618 | 2,192 | +35% | 0 | 0 | — |
case-10 | fail→fail | 11,072 | 11,790 | +6% | 1 | 1 | 0% | 1,547 | 2,861 | +85% | 0 | 0 | — |
case-11 | fail→fail | 9,090 | 7,940 | -13% | 1 | 1 | 0% | 1,382 | 2,199 | +59% | 0 | 0 | — |
case-12 | fail→fail | 10,775 | 5,122 | -52% | 1 | 1 | 0% | 1,915 | 1,927 | +1% | 0 | 0 | — |
case-13 | fail→pass | 9,183 | 7,385 | -20% | 1 | 1 | 0% | 1,328 | 2,158 | +63% | 0 | 0 | — |
case-14 | fail→fail | 10,210 | 7,437 | -27% | 1 | 1 | 0% | 1,431 | 2,203 | +54% | 0 | 0 | — |
case-15 | pass→pass | 9,061 | 8,837 | -2% | 1 | 1 | 0% | 1,354 | 2,357 | +74% | 0 | 0 | — |
case-16 | fail→pass | 13,516 | 9,934 | -27% | 1 | 1 | 0% | 1,979 | 2,614 | +32% | 0 | 0 | — |
case-17 | pass→pass | 15,385 | 9,259 | -40% | 1 | 1 | 0% | 2,263 | 2,543 | +12% | 0 | 0 | — |
case-18 | fail→fail | 8,866 | 9,454 | +7% | 1 | 1 | 0% | 1,289 | 2,567 | +99% | 0 | 0 | — |
case-19 | pass→pass | 10,505 | 7,350 | -30% | 1 | 1 | 0% | 1,616 | 2,195 | +36% | 0 | 0 | — |
case-20 | fail→fail | 9,662 | 4,297 | -56% | 1 | 1 | 0% | 1,477 | 1,688 | +14% | 0 | 0 | — |
case-21 | fail→fail | 10,953 | 6,404 | -42% | 1 | 1 | 0% | 1,661 | 2,035 | +23% | 0 | 0 | — |
case-22 | fail→fail | 18,193 | 11,695 | -36% | 1 | 1 | 0% | 2,795 | 2,919 | +4% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +27 percentage points is the difference between those two pass rates over the 22 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.