Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Lap plan content ky sau TU DATA ky truoc — doc audit + report, giu winner, thay loser, thu nghiem moi theo ti le 70/20/10, toi da 2 hypothesis moi/ky. Khong lap plan tu cam tinh. Kich hoat khi nhac 'next content plan', 'plan content thang toi', 'ke hoach content dua tren data', 'content ky sau'.
.claude/skills/minhnv0807-40-next-content-plan/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 37% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 83% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 8% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 81% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 81% | 0% |
> Plan khong co data cu = doan mo. Thu tu cung: quet data → nhan winner → test hypothesis moi → len lich. Khong thay doi tat ca cung luc — ky nao cung dap di xay lai la khong bao gio biet dieu gi work.
Truoc khi viet bat ky dong plan nao, can co:
39-content-audit) — neu chua co, chay 39 truoc, khong lap plan chay.07-bao-cao-marketing) neu co.Neu user khong cung cap duoc data toi thieu (audit hoac danh sach bai + metric), dung lai va de xuat chay 39-content-audit truoc.
01-lich-noi-dung.Tu audit (39) + report (07), dien bang 3 cot:
| Da work (GIU) | Khong work (DUNG) | Chua thu (TEST MOI) | |---------------|-------------------|---------------------| | format/angle/kenh + so lieu chung minh] | + ly do tu data] | y tuong tu pattern hoac insight moi] |
| Ky lap plan | Muc tieu chinh | KPI uu tien | Campaign kem | Thay doi lon | |-------------|----------------|-------------|--------------|--------------| | | | | | |
Muc tieu chinh chi 1 — reach / lead / conversion / retention. KPI uu tien la metric se dung de danh gia ky nay.
| Nhom | Ti le | Noi dung | |------|-------|----------| | Proven (70%) | ~70% so bai | Winner giu nguyen cong thuc — chi thay chu de | | Toi uu (20%) | ~20% so bai | Bien the tu winner: doi hook / kenh / do dai | | Thu nghiem (10%) | ~10% so bai | Hypothesis moi (toi da 2) |
Kem bang dieu chinh ti le pheu:
| Loai | Ti le ky truoc | Ti le ky moi | Ly do thay doi (tu data) | |------|----------------|--------------|--------------------------| | TOFU / MOFU / BOFU | | | | | Video vs Static | | | |
| Winner goc | Ket qua (metric) | Nhan ban thanh | Thay doi gi | Kenh / Thoi diem | |------------|------------------|----------------|-------------|------------------| | | | | | |
Moi winner nhan thanh 2-3 bien the — giu cau truc hook + format, thay chu de / goc quay / nhan vat.
| Hypothesis | Format | Kenh | KPI do | Thoi gian test | Tieu chi ket luan | |-----------|--------|------|--------|----------------|-------------------| | | | | | | |
Hypothesis phai viet dang kiem chung duoc: "Neu thay doi X] thi metric Y] se tang/giam Z%]". Setup test chi tiet → goi 19-ab-test-setup.
| Tuan | Theme | Kenh uu tien | So bai | Highlight (winner nhan ban / hypothesis) | |------|-------|--------------|--------|------------------------------------------| | Tuan 1 | | | | | | Tuan 2 | | | | | | Tuan 3 | | | | | | Tuan 4 | | | | |
| Nguoi | Trach nhiem | So bai/tuan | Kenh chinh | |-------|-------------|-------------|------------| | | | | |
Chi tiet tung ngay (gio dang, hook, CTA, trang thai) → chay 01-lich-noi-dung. Brief tung bai → 36-content-brief.
Ten file: next-content-plan-[ten-san-pham]-[YYYYMMDD].md
markdown# Next Content Plan — [Ky] — [San pham/Brand] ## I. Review ky truoc | Da work (giu) | Khong work (dung) | Chua thu (test moi) | |---------------|-------------------|---------------------| ## II. Dinh huong ky moi | Ky lap plan | Muc tieu chinh | KPI uu tien | Campaign kem | Thay doi lon | ## III. Content mix ky moi — 70/20/10 | Nhom | Ti le | So bai | Noi dung | [+ bang dieu chinh TOFU/MOFU/BOFU, video vs static — kem ly do tu data] ## IV. Nhan winner | Winner goc | Ket qua | Nhan ban thanh | Thay doi gi | Kenh / Thoi diem | ## V. Hypothesis moi (toi da 2) | Hypothesis | Format | Kenh | KPI do | Thoi gian test | Tieu chi ket luan | ## VI. Lich tong quan | Tuan | Theme | Kenh uu tien | So bai | Highlights | ## VII. Phan cong | Nguoi | Trach nhiem | So bai/tuan | Kenh chinh |
39-content-audit: input bat buoc — chay truoc plan nay. Khong co audit = khong co plan.07-bao-cao-marketing: report ky truoc bo sung goc nhin KPI tong.01-lich-noi-dung: nhan plan nay va trien khai lich chi tiet tung ngay.36-content-brief: brief tung bai theo plan.19-ab-test-setup: setup test cho 2 hypothesis moi.13-phan-tich-du-lieu: ho tro doc data khi can dao sau.01-lich-noi-dung cho lich chi tiet va 19-ab-test-setup cho hypothesis| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 36,150 | 39,949 | +11% | 1 | 1 | 0% | 5,863 | 8,041 | +37% | 0 | 0 | — |
case-02 | fail→pass | 16,497 | 15,893 | -4% | 1 | 1 | 0% | 2,175 | 3,985 | +83% | 0 | 0 | — |
case-03 | fail→pass | 43,545 | 31,096 | -29% | 1 | 1 | 0% | 6,271 | 6,800 | +8% | 0 | 0 | — |
case-04 | pass→pass | 24,806 | 26,259 | +6% | 1 | 1 | 0% | 2,940 | 5,460 | +86% | 0 | 0 | — |
case-05 | pass→pass | 25,988 | 20,874 | -20% | 1 | 1 | 0% | 3,374 | 5,223 | +55% | 0 | 0 | — |
case-06 | pass→pass | 28,771 | 32,617 | +13% | 1 | 1 | 0% | 4,751 | 6,857 | +44% | 0 | 0 | — |
case-07 | fail→fail | 30,394 | 25,850 | -15% | 1 | 1 | 0% | 3,831 | 5,793 | +51% | 0 | 0 | — |
case-08 | fail→pass | 19,567 | 28,046 | +43% | 1 | 1 | 0% | 3,210 | 5,814 | +81% | 0 | 0 | — |
case-09 | fail→pass | 22,453 | 21,396 | -5% | 1 | 1 | 0% | 3,132 | 5,657 | +81% | 0 | 0 | — |
case-10 | pass→fail | 18,334 | 26,450 | +44% | 1 | 1 | 0% | 3,116 | 5,320 | +71% | 0 | 0 | — |
case-11 | pass→pass | 19,799 | 27,009 | +36% | 1 | 1 | 0% | 2,551 | 5,486 | +115% | 0 | 0 | — |
case-12 | fail→pass | 16,443 | 22,304 | +36% | 1 | 1 | 0% | 2,868 | 5,019 | +75% | 0 | 0 | — |
case-13 | fail→pass | 23,119 | 29,004 | +25% | 1 | 1 | 0% | 2,869 | 6,111 | +113% | 0 | 0 | — |
case-14 | fail→fail | 15,900 | 10,211 | -36% | 1 | 1 | 0% | 1,710 | 2,810 | +64% | 0 | 0 | — |
case-15 | pass→pass | 22,740 | 21,219 | -7% | 1 | 1 | 0% | 2,896 | 4,926 | +70% | 0 | 0 | — |
case-16 | fail→pass | 17,801 | 10,614 | -40% | 1 | 1 | 0% | 1,860 | 2,878 | +55% | 0 | 0 | — |
case-17 | fail→pass | 23,286 | 11,878 | -49% | 1 | 1 | 0% | 3,028 | 3,202 | +6% | 0 | 0 | — |
case-18 | fail→fail | 25,870 | 22,824 | -12% | 1 | 1 | 0% | 3,484 | 4,879 | +40% | 0 | 0 | — |
case-19 | pass→pass | 31,043 | 20,999 | -32% | 1 | 1 | 0% | 2,986 | 5,323 | +78% | 0 | 0 | — |
case-20 | pass→pass | 17,309 | 25,172 | +45% | 1 | 1 | 0% | 2,883 | 5,223 | +81% | 0 | 0 | — |
case-21 | fail→pass | 25,291 | 21,044 | -17% | 1 | 1 | 0% | 3,235 | 4,702 | +45% | 0 | 0 | — |
case-22 | fail→pass | 19,408 | 21,470 | +11% | 1 | 1 | 0% | 3,097 | 4,664 | +51% | 0 | 0 | — |
case-23 | fail→pass | 20,705 | 16,062 | -22% | 1 | 1 | 0% | 1,806 | 3,844 | +113% | 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. 23 cases were attempted. The headline lift of +48 percentage points is the difference between those two pass rates over the 23 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.