Skip to content

yanliudesign/offer-toolkit-skill / salary-negotiation-skill

salary-negotiation-skill

Verified

frameworks/negotiation-tactics.md

View on GitHub
# Negotiation Tactics

一份**战术库**。生成脚本时选用;报告里可以引用("这里用的是 anchoring 战术")。

---

## 战术 1 · Anchor High

**原理**:先出手的一方设定谈判 frame。第一个数字往往是整个谈判的重力锚。

**用法**:
- 你的 target 是 $220K base → 你不说 "I want $220K",你说 "I'm looking at $225-230K based on…"
- Recruiter 会 counter 回来,通常落在你的 $220 附近 = 你的原目标

**误区**:Anchor 得太离谱(比如 asking $280K 当 P50 是 $220K)会让 recruiter 觉得你不懂市场,谈判可信度崩塌。**Anchor 应该高于 target 5-10%,不是 30%。**

**报告里的语言**:不要写 "anchor tactic" 这种术语,直接给数字("target ask 是 $225-230K")。战术自己 encode 在数字里。

---

## 战术 2 · Silence(沉默)

**原理**:招聘官说完话后停顿 3-5 秒,会 90% 概率继续说话,往往把 concession 补上("or I could see if there's something on sign-on…")。

**用法**:
- 电话里:说完你的 ask,闭嘴。不要 "and I know it's a lot to ask, but…" 这种 filler
- 邮件里:等 recruiter 回复,不要 24 小时内 nudge

**用户常见错误**:紧张时不停解释,把 leverage 说没了。**电话脚本 Beat 4 之后必须写 "STOP TALKING"。**

---

## 战术 3 · Reframe(重新定义)

**原理**:招聘官抛出一个不利 frame("this is our best offer"),你不接受这个 frame,重新定义讨论对象。

**例子**:

Recruiter: "This is our best offer."
❌ You: "Are you sure? Can you go higher?"(accepts frame)
✅ You: "When you say 'best,' is that the total package or specifically base? Because different components usually have different approvers."(reframes)

Recruiter: "Budget is fixed."
❌ You: "Can budget be moved?"(accepts)
✅ You: "Understood. Given budget constraints, are there non-cash levers — early review, refresh commitment, or a start-date bonus?"(reframes)

Recruiter: "We may need to reconsider the pool."
❌ You: "I'll accept whatever you can offer."(capitulates)
✅ You: "I hear you. I want to be equally direct — I'm not negotiating against a threat, I'm negotiating against market data. If the numbers don't work, I understand. But I'd rather have that conversation now."(reframes: threat → data conversation)

---

## 战术 4 · Package vs Component

**原理**:招聘官思考"总 TC 达 X",你可以要 "same TC, different mix"(更多 sign-on 换更少 RSU 等)。

**用法**:谈判后期 recruiter 卡在总数字 → 提出 "same total, more base less RSU"(如果你风险偏好保守)或 "same total, more RSU less base"(如果你 bullish 公司股价)

**报告里怎么用**:Modality A 卡在第 2 轮时的 backup ask 方案。

---

## 战术 5 · The Third Option(第三选项)

**原理**:当 recruiter 说 "we can do A or B, pick one",你**发明 C**。

**例子**:Recruiter: "We can either bump base by $5K or sign-on by $15K, not both."
✅ You: "Instead of either, could we do +$3K base + $10K sign-on? Same $13K total but split, and it helps me on both Year 1 cash and long-term run-rate."

**为什么 work**:让 recruiter 感觉他还是 "picking",但你拿到了 blended outcome。**很多招聘官被这招 charmed** —— 因为你在 solve his problem,不是要 more from him。

---

## 战术 6 · The Concession Trade

**原理**:不要免费给 concession。招聘官问 "can you start earlier?" —— 你答 "yes, if you can add X"。

**常见 trade**:
- 早 start → 换 sign-on 加 $10-20K
- 签更长的 clawback 窗口(24 vs 12 months) → 换 sign-on 加 30%
- 接受较低 base → 换 RSU 加 15%+
- 放弃 relocation lump sum(自己搬) → 换 sign-on 加 relocation cost 的 1.5x

---

## 战术 7 · Name the Emotion

**原理**:招聘官用情绪 pressure("I really pushed for you, this hurts me")时,中性 acknowledge,不 counter-emotion。

**用法**:
- Recruiter: "I really went to bat for you, and this is what came back."
- ✅ You: "I hear that, and I appreciate it. I don't doubt you're advocating. What I'm asking for isn't personal — it's about the market data. If the answer is genuinely no, I understand, but I want to make sure the ask is on the table for internal reference."
- ❌ You: "I'm sorry if this is uncomfortable."(apologize = 认错 = 消 leverage)

---

## 战术 8 · The 24-Hour Rule

**原理**:任何"unexpected"数字(比你想象高或低)都不当场答,24 小时后回。

**理由**:
- 大数字:你可能 undervalue,需要重新算 TC 或 check competing offer
- 小数字:情绪反应可能让你 impulsively reject 或 accept
- Recruiter 用"当天答"制造 pressure = 强 signal 他自己也在压力下 = 你有 leverage

**脚本**:
> "This is a big decision, and I want to give it the thought it deserves. Let me sleep on it and come back to you tomorrow morning."

Recruiter 拒绝这句 = 罕见 = 你重新评估 offer quality。

---

## 战术 9 · The Bracket

**原理**:报数字时永远给 range,target 数字接近 range 中间偏低。

**为什么**:Recruiter 会往 range 底部走。你的 target 在中间 = 谈到你的 target 是"胜利"。

**例子**:Target base $220K → Ask "$225-230K" → Recruiter counters $220K → 你的原 target。

---

## 战术 10 · The Multi-Anchor

**原理**:给 recruiter 多个 anchor 让他去 anchor 你想要的那个。

**例子**:谈 RSU
> "Comparable roles at [Company X] are around $700K equity, and at [Company Y] around $780K. Given [Current Company]'s position in the market, something in the $720-750K range would feel right."

**Recruiter 会 anchor $720**(最低你给的 range),但你已经把讨论从"当前 $600K"pulled to "$720-780K bracket"。

---

## 战术 11 · The Written Confirmation

**原理**:任何口头承诺(refresh、early review、bonus guarantee)必须在邮件里写下来。

**脚本**:
> "As we discussed on the call, I want to confirm in writing so we have a shared record: [list items]. Please let me know if my understanding is accurate, and I'll consider this the final package for signature."

**Recruiter 如果 back off**("I didn't quite say that")→ 你回:"Ah OK — I want to make sure we're aligned before I sign. Can you clarify what was said? I'll wait for confirmation."

**这一步能 catch 80% 的口头承诺 walking back**。

---

## 战术 12 · The Silent Bluff Test

**原理**:招聘官说 "reconsider the pool" 时,你的沉默 24 小时是 bluff-detector。

**用法**:
- 收到威胁邮件 → 24 小时不回
- 如果 recruiter 主动 follow up 说 "just checking in" → **他在 bluff**,你 leverage 反而回来了
- 如果 recruiter 也 24 小时沉默 → 可能是真的 → 你重新评估要不要 pull back

---

## 关键 anti-pattern(脚本中禁止出现)

| 禁用短语 | 为什么 | 替代 |
|---|---|---|
| "I really need…" | 暴露 desperation | "I'd like to see…" |
| "Is that the best you can do?" | 空 pushback | "Given [specific reason], can we explore [specific ask]?" |
| "I don't want to be difficult, but…" | 提前 apologize = 认错 | 直接说事 |
| "This is my final ask" | 你 lock 自己 | Recruiter 会 test |
| "I have to have $X" | Ultimatum | "Based on [data], I need to land around $X to feel comfortable" |
| "Just curious what's possible…" | 太 tentative | 直接给数字 |
| "Sorry to bother, but…" | Weak opener | 直接进入 |
| "I understand if you can't" (Round 1) | 太早 concede | 留到最后 walk-away 时用 |

---

## 关键 pattern(脚本必须常用)

| Pattern | 例子 |
|---|---|
| **"Given [data]…"** | "Given Levels.fyi P50 is $X…" |
| **"I'd love to see…"** | "I'd love to see base come up to…" |
| **"Help me understand…"** | "Help me understand where the flex is vs. what's fixed." |
| **"If X, then Y"** | "If base is capped, is equity separately approvable?" |
| **"I want to be direct — [message]"** | Signals honesty, disarms |
| **"I appreciate you [action]"** | Preserves relationship |
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 322kUpdated Claude CodeCodexVerifiedInstall