Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →This skill enables Perplexity AI to identify, research, and qualify potential sales prospects, building targeted prospect lists and personalized outreach intelligence to accelerate pipeline development.
.claude/skills/vythanhtra-skill-sales-prospecting/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-11 | ✗→✓ | ▲ Improved | 69% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 42% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 16% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 83% | 0% |
| case-17 | ✗→✓ | ▲ Improved | 115% | 0% |
This skill enables Perplexity AI to identify, research, and qualify potential sales prospects, building targeted prospect lists and personalized outreach intelligence to accelerate pipeline development.
yamlrequired_inputs: - product_service: "What you are selling" - icp_criteria: "Ideal customer profile definition" - prospecting_goal: "List building|Single account research|Contact identification" optional_inputs: - target_industry: "Specific industries to focus on" - company_size: "Employee count range" - geography: "Target regions or markets" - persona: "Job titles and roles to target" - trigger_events: "Signals to prioritize (funding, hiring, news)" - competitors: "Competitors whose customers to target" - exclusion_list: "Companies or domains to exclude"
Step 1: ICP Criteria Documentation
Step 2: ICP Validation
Step 3: Company Discovery
Search strategies:
Sources to mine:
Step 4: Contact Identification
Buying committee roles to find:
Step 5: Account Deep-Dive
Step 6: Buying Signal Detection
| Signal Type | What to Look For | Priority | |------------|-----------------|----------| | Funding | New funding round | High | | Leadership | New C-suite hire | High | | Growth | Rapid hiring in target dept | High | | Expansion | New market or product launch | Medium | | Pain Signal | Negative reviews of competitor | Medium | | Content | Published content on your topic | Low |
Step 7: Personalization Research
Step 8: Outreach Brief Creation For each prospect:
# Prospect List: [Campaign Name]
Created: [Date] | ICP: [Brief description] | Goal: [Number]
## Tier 1 Prospects (Best Fit)
| Company | Contact | Title | Signal | Score |
|---------|---------|-------|--------|-------|
## Tier 2 Prospects (Good Fit)
| Company | Contact | Title | Signal | Score |
|---------|---------|-------|--------|-------|# Account Profile: [Company Name]
Researched: [Date] | Tier: 1/2/3 | Owner: [Rep Name]
## Company Snapshot
- Industry: [Sector] | Size: [Employees] | Stage: [Startup/Growth/Enterprise]
- Website: [URL]
## Why They Fit Our ICP
- [Specific ICP criterion 1 and evidence]
- [Technology or behavior signal]
## Recent Intelligence
- [Most recent relevant news or development]
- [Known challenge or pain point]
## Buying Signals
- [Signal 1]: [Details and date]
## Key Contacts
### [Name] - [Title]
- Role in buying: [Economic/Champion/User/Technical]
- Background: [2-3 sentences]
- Outreach hook: [Personalization angle]
## Proposed Outreach
- Channel: [Email/Phone/LinkedIn]
- Subject: [Draft subject line]
- Hook: [Personalized first line]
- Value prop: [1-2 sentences tailored to their context]
- CTA: [Specific next step]yamlscoring_criteria: firmographic_fit: industry_match: 20 points size_match: 15 points geography_match: 10 points technographic_fit: complementary_tools: 15 points replaces_competitor_tool: 20 points behavioral_signals: recent_funding: 15 points leadership_change: 10 points active_hiring: 10 points content_engagement: 5 points scoring_tiers: tier_1: 70-100 points (immediate outreach) tier_2: 40-69 points (nurture sequence) tier_3: 20-39 points (monitor and revisit) disqualified: below 20 points
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-11 | fail→pass | 10,170 | 4,193 | -59% | 1 | 1 | 0% | 1,618 | 2,732 | +69% | 0 | 0 | — |
case-01 | fail→fail | 34,677 | 23,508 | -32% | 1 | 1 | 0% | 5,544 | 5,821 | +5% | 0 | 0 | — |
case-02 | fail→fail | 29,287 | 23,985 | -18% | 1 | 1 | 0% | 4,960 | 6,026 | +21% | 0 | 0 | — |
case-03 | fail→fail | 18,678 | 17,789 | -5% | 1 | 1 | 0% | 3,143 | 5,475 | +74% | 0 | 0 | — |
case-04 | fail→fail | 14,187 | 22,427 | +58% | 1 | 1 | 0% | 2,512 | 4,837 | +93% | 0 | 0 | — |
case-05 | fail→fail | 22,205 | 11,138 | -50% | 1 | 1 | 0% | 2,160 | 3,787 | +75% | 0 | 0 | — |
case-06 | fail→pass | 11,751 | 10,397 | -12% | 1 | 1 | 0% | 1,996 | 2,842 | +42% | 0 | 0 | — |
case-07 | pass→pass | 8,981 | 9,648 | +7% | 1 | 1 | 0% | 1,581 | 3,673 | +132% | 0 | 0 | — |
case-08 | pass→fail | 13,335 | 12,418 | -7% | 1 | 1 | 0% | 2,114 | 4,102 | +94% | 0 | 0 | — |
case-09 | fail→fail | 6,001 | 3,084 | -49% | 1 | 1 | 0% | 915 | 2,542 | +178% | 0 | 0 | — |
case-10 | fail→pass | 12,763 | 1,556 | -88% | 1 | 1 | 0% | 2,001 | 2,313 | +16% | 0 | 0 | — |
case-12 | pass→pass | 13,507 | 12,821 | -5% | 1 | 1 | 0% | 2,327 | 4,399 | +89% | 0 | 0 | — |
case-13 | fail→pass | 10,899 | 8,199 | -25% | 1 | 1 | 0% | 1,926 | 3,529 | +83% | 0 | 0 | — |
case-14 | pass→fail | 14,114 | 6,646 | -53% | 1 | 1 | 0% | 2,421 | 3,064 | +27% | 0 | 0 | — |
case-15 | pass→pass | 9,193 | 3,740 | -59% | 1 | 1 | 0% | 1,757 | 2,567 | +46% | 0 | 0 | — |
case-16 | pass→pass | 9,497 | 2,742 | -71% | 1 | 1 | 0% | 1,496 | 2,431 | +63% | 0 | 0 | — |
case-17 | fail→pass | 11,647 | 13,075 | +12% | 1 | 1 | 0% | 1,966 | 4,229 | +115% | 0 | 0 | — |
case-18 | fail→pass | 6,957 | 2,236 | -68% | 1 | 1 | 0% | 1,215 | 2,450 | +102% | 0 | 0 | — |
case-19 | fail→fail | 16,525 | 15,013 | -9% | 1 | 1 | 0% | 2,892 | 4,687 | +62% | 0 | 0 | — |
case-20 | fail→fail | 17,189 | 17,248 | +0% | 1 | 1 | 0% | 3,606 | 5,819 | +61% | 0 | 0 | — |
case-21 | fail→fail | 20,618 | 26,978 | +31% | 1 | 1 | 0% | 3,590 | 6,967 | +94% | 0 | 0 | — |
case-22 | fail→pass | 9,666 | 2,150 | -78% | 1 | 1 | 0% | 1,656 | 2,454 | +48% | 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 +23 percentage points is the difference between those two pass rates over the 22 comparable cases. 2 cases got worse with the skill loaded, and they are included in that figure.
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.