Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →When the user wants to plan email marketing, EDM, newsletter strategy, or email deliverability. Also use when the user mentions "email marketing," "EDM," "newsletter," "SPF," "DKIM," "DMARC," "email deliverability," "email content strategy," "email campaigns," "newsletter strategy," "email automation," or "cold email." For signup UI, use newsletter-signup-generator.
.claude/skills/mkurman-email-marketing/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 39% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 6% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 29% | 0% |
| case-15 | ✗→✓ | ▲ Improved | 64% | 0% |
| case-19 | ✗→✓ | ▲ Improved | 38% | 0% |
|---------|-----| | EDM | Direct marketing; conversion-focused | Promotions, campaigns, announcements; bulk sends | | Newsletter | Ongoing value; relationship | Industry insights, curated articles; regular cadence |
Combine both: EDM for push; Newsletter for nurture. Cover different stages and goals.
| Type | Use | |------|-----| | Onboarding | Welcome + first-use guidance; 5-7 day sequence; behavior-triggered; drive "Aha!" moment | | Campaign | Promotions, limited-time; conversion or participation | | Announcement | Product launch, major update; one-time important notice | | Features update | New features, improvements; help users adopt | | Blog/Newsletter | Curated articles, industry insights; sustained touch |
| Practice | Guideline | |----------|-----------| | Personalization | Segment by behavior, source, stage; boosts open/click | | Timing | New users: dense; existing: controlled pace; behavior-triggered > calendar-only | | Welcome series | Send soon after signup; 5-7 emails over days; guide first key action | | Unsubscribe | One-click required (Gmail/Yahoo); honor within 48h; clear entry | | Complaint rate | Keep below 0.3%; list hygiene critical |
| Practice | Guideline | |----------|-----------| | Subject | One clear topic per email; avoid pure promo | | Value first | Useful info before promotion | | CTA | Single primary CTA; clear next step | | Mobile | 50%+ read on mobile; responsive layout, tappable links |
Subdomain: Use subdomain (e.g. mail.example.com) for marketing; keep transactional (support@, etc.) on main domain. Isolate risk.
| Protocol | Purpose | |----------|---------| | SPF | Authorizes mail servers for domain | | DKIM | Cryptographic signature; verifies sender | | DMARC | Policy for unauthenticated mail; start p=none, then quarantine, then reject over 60-90 days |
Order: SPF first, then DKIM, then DMARC. Gmail/Yahoo require all three for bulk senders (5,000+/day) since Feb 2024.
Advanced: TLS-RPT, MTA-STS, BIMI (brand logo). Postmaster Tools: Monitor deliverability, spam rate, auth status.
| Article Type | Use | |--------------|-----| | Retention | Deep content for existing users; improve retention | | ToFu | Top of funnel; awareness (trends, concepts, problem framing) | | MoFu | Middle of funnel; consideration (comparisons, reviews, best practices) |
Dual value: (1) Better email engagement (open, click, stickiness); (2) Drive traffic to article pages from non-search channel; signals to Google that users value content; supports SEO.
Measurement: GA4 email source traffic to article pages; GSC rank/click changes.
| Guideline | Note | |-----------|------| | Baseline | 1 high-value email/week for most brands | | Peak times | Tue-Thu, 8-11am or 2-4pm (recipient timezone) | | Segmentation | New vs loyal need different cadence | | Quality | Relevant, behavior-triggered > calendar volume |
Data: ~36% send 1-3/month; ~30% weekly; daily risks high unsubscribe.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 33,732 | 34,898 | +3% | 1 | 1 | 0% | 5,082 | 7,058 | +39% | 0 | 0 | — |
case-02 | fail→fail | 31,418 | 25,445 | -19% | 1 | 1 | 0% | 4,578 | 5,335 | +17% | 0 | 0 | — |
case-03 | fail→pass | 38,295 | 26,093 | -32% | 1 | 1 | 0% | 5,279 | 5,583 | +6% | 0 | 0 | — |
case-04 | pass→pass | 15,622 | 16,258 | +4% | 1 | 1 | 0% | 2,630 | 3,818 | +45% | 0 | 0 | — |
case-05 | fail→pass | 16,212 | 16,331 | +1% | 1 | 1 | 0% | 2,892 | 3,725 | +29% | 0 | 0 | — |
case-06 | pass→pass | 11,388 | 8,961 | -21% | 1 | 1 | 0% | 1,886 | 2,633 | +40% | 0 | 0 | — |
case-07 | fail→fail | 15,982 | 15,217 | -5% | 1 | 1 | 0% | 2,469 | 3,516 | +42% | 0 | 0 | — |
case-08 | pass→pass | 14,557 | 9,880 | -32% | 1 | 1 | 0% | 2,115 | 2,867 | +36% | 0 | 0 | — |
case-09 | pass→pass | 13,554 | 20,048 | +48% | 1 | 1 | 0% | 2,155 | 3,360 | +56% | 0 | 0 | — |
case-10 | pass→pass | 5,443 | 7,805 | +43% | 1 | 1 | 0% | 845 | 2,357 | +179% | 0 | 0 | — |
case-11 | fail→fail | 13,067 | 11,123 | -15% | 1 | 1 | 0% | 2,021 | 2,764 | +37% | 0 | 0 | — |
case-12 | pass→pass | 14,436 | 10,030 | -31% | 1 | 1 | 0% | 2,163 | 2,568 | +19% | 0 | 0 | — |
case-13 | pass→pass | 12,440 | 12,590 | +1% | 1 | 1 | 0% | 2,004 | 3,064 | +53% | 0 | 0 | — |
case-14 | pass→pass | 15,083 | 16,653 | +10% | 1 | 1 | 0% | 2,363 | 3,913 | +66% | 0 | 0 | — |
case-15 | fail→pass | 13,119 | 14,954 | +14% | 1 | 1 | 0% | 2,185 | 3,588 | +64% | 0 | 0 | — |
case-16 | pass→pass | 16,220 | 74,465 | +359% | 1 | 1 | 0% | 2,371 | 4,322 | +82% | 0 | 0 | — |
case-17 | pass→pass | 15,584 | 17,428 | +12% | 1 | 1 | 0% | 2,396 | 3,733 | +56% | 0 | 0 | — |
case-18 | fail→fail | 14,212 | 16,263 | +14% | 1 | 1 | 0% | 2,277 | 3,809 | +67% | 0 | 0 | — |
case-19 | fail→pass | 15,646 | 14,523 | -7% | 1 | 1 | 0% | 2,492 | 3,428 | +38% | 0 | 0 | — |
case-20 | pass→pass | 25,310 | 14,551 | -43% | 1 | 1 | 0% | 5,364 | 4,124 | -23% | 0 | 0 | — |
case-21 | pass→pass | 13,492 | 13,283 | -2% | 1 | 1 | 0% | 2,360 | 3,455 | +46% | 0 | 0 | — |
case-22 | pass→pass | 12,223 | 14,937 | +22% | 1 | 1 | 0% | 1,910 | 3,329 | +74% | 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. 1 case got worse with the skill loaded, and it is 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.