Install any skill in seconds. Free to start, no credit card required.
Get Started Free →运用《销售巨人》(Neil Rackham) 的 SPIN 提问序列与晋级承诺方法推进大订单/B2B 复杂销售。适用于用户要准备大客户拜访或售前会谈、设计销售提问话术、诊断"聊得很好但单子不动"的僵局、以及规划多轮销售的阶段目标时使用。与 persuasion-pitch(框架控制与融资路演)互补:pitch 管气场与叙事,SPIN 管提问与需求挖掘。
.claude/skills/kuhung-spin-selling/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-11 | ✗→✓ | ▲ Improved | 35% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 29% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 5% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 15% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 63% | 0% |
你是一名基于实证研究的 B2B 销售教练。你的使命是帮助用户戒掉"介绍产品 + 逼单"的小订单打法,改用 SPIN 提问序列让客户自己说出购买价值,并用晋级承诺推进多轮销售。
详细论据与研究数据见 notes/销售巨人_笔记.md。
暂无。技能在实战中暴露的偏差会以 - YYYY-MM-DD: 经验内容 格式追加到本章节。
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-03 | fail→fail | 15,966 | 14,531 | -9% | 1 | 1 | 0% | 2,260 | 2,902 | +28% | 0 | 0 | — |
case-04 | fail→fail | 19,112 | 17,617 | -8% | 1 | 1 | 0% | 2,774 | 3,278 | +18% | 0 | 0 | — |
case-11 | fail→pass | 17,742 | 18,570 | +5% | 1 | 1 | 0% | 2,463 | 3,325 | +35% | 0 | 0 | — |
case-12 | fail→pass | 17,969 | 17,288 | -4% | 1 | 1 | 0% | 2,628 | 3,377 | +29% | 0 | 0 | — |
case-13 | pass→pass | 17,977 | 14,759 | -18% | 1 | 1 | 0% | 2,645 | 3,071 | +16% | 0 | 0 | — |
case-14 | pass→pass | 38,300 | 19,721 | -49% | 1 | 1 | 0% | 2,676 | 3,670 | +37% | 0 | 0 | — |
case-15 | pass→pass | 18,598 | 14,269 | -23% | 1 | 1 | 0% | 2,614 | 2,829 | +8% | 0 | 0 | — |
case-01 | fail→fail | 20,151 | 20,234 | +0% | 1 | 1 | 0% | 2,794 | 3,988 | +43% | 0 | 0 | — |
case-02 | fail→pass | 23,258 | 18,346 | -21% | 1 | 1 | 0% | 3,278 | 3,438 | +5% | 0 | 0 | — |
case-05 | pass→pass | 17,936 | 15,518 | -13% | 1 | 1 | 0% | 2,658 | 3,245 | +22% | 0 | 0 | — |
case-06 | pass→pass | 22,101 | 18,022 | -18% | 1 | 1 | 0% | 2,984 | 3,455 | +16% | 0 | 0 | — |
case-07 | pass→pass | 18,451 | 17,124 | -7% | 1 | 1 | 0% | 2,565 | 3,280 | +28% | 0 | 0 | — |
case-08 | fail→pass | 20,986 | 16,576 | -21% | 1 | 1 | 0% | 2,792 | 3,200 | +15% | 0 | 0 | — |
case-09 | fail→pass | 10,778 | 12,820 | +19% | 1 | 1 | 0% | 1,618 | 2,634 | +63% | 0 | 0 | — |
case-10 | pass→pass | 9,581 | 12,043 | +26% | 1 | 1 | 0% | 1,493 | 2,727 | +83% | 0 | 0 | — |
case-16 | pass→pass | 20,287 | 17,225 | -15% | 1 | 1 | 0% | 2,966 | 3,351 | +13% | 0 | 0 | — |
case-17 | pass→pass | 16,690 | 17,767 | +6% | 1 | 1 | 0% | 2,307 | 3,519 | +53% | 0 | 0 | — |
case-18 | pass→pass | 15,644 | 16,318 | +4% | 1 | 1 | 0% | 2,196 | 3,193 | +45% | 0 | 0 | — |
case-19 | pass→pass | 17,812 | 21,987 | +23% | 1 | 1 | 0% | 2,678 | 3,881 | +45% | 0 | 0 | — |
case-20 | pass→pass | 21,508 | 19,569 | -9% | 1 | 1 | 0% | 3,176 | 3,712 | +17% | 0 | 0 | — |
case-21 | pass→pass | 20,533 | 17,699 | -14% | 1 | 1 | 0% | 2,861 | 3,399 | +19% | 0 | 0 | — |
case-22 | pass→pass | 19,171 | 20,854 | +9% | 1 | 1 | 0% | 3,094 | 4,077 | +32% | 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.
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.