「王总，是不是方案哪里不合适？如果预算有压力我们还能再谈。」你把「他嫌贵」这个纯属想象的原因送到他面前，还顺手降了价。真实原因若是财务还没批，你白送了一轮折扣，还教会他：只要不回，价格就会自己往下掉。
「王总，你那边一直没消息，我猜不出是排不上号、内部卡住了、还是方向变了——这三种情况我该做的事完全不一样。你回个数字就行：1 排不上号 / 2 内部卡住 / 3 不做了。」
—— 不解码、不脑补、不自降价，只把「告诉我你在哪一格」压到一个字符。
EN Silence is not a signal, it's a default: nothing happened. One silence is simultaneously compatible with six different situations — buried inbox, needs to ask someone first, internal opposition, budget slipped, already chose someone else, genuinely not interested — and the right response to each contradicts the others. So the information content of silence is close to zero. The problem is that people cannot tolerate a blank, so they fill it with the most self-relevant reading available: they think I'm too expensive, they're sick of me. Two well-documented biases push the same way. The spotlight effect means you badly overestimate how much space you occupy in their head; and in text, senders systematically overrate how well their tone comes across. So the honest move is to stop decoding and hand the decoding back — at a price of one keystroke.
机制：人脑不容忍空白，会用最可得的解释填上——而对你最可得的永远是跟你有关的那种。你既高估了自己在他脑子里的分量，也高估了那条消息传出去的语气。两者叠加，沉默被自动翻译成「他对我有意见」。
再加一层 Ross (1977) 的基本归因错误：你把处境读成人品——他不回最可能是「今天有十七件更急的事」，你却读成「他不想理我」。但沉默偶尔确实是信号，判据是基线：十分钟必回的人在他自己定的日期后消失，那是变化；本就三天回一次的人隔五天没回，什么都不是。也别读成「不必反省」——沉默名单越来越长，那是你的东西本就不重要。
"I'd rather ask than assume." —— 承认在猜，比猜错便宜。
"Reply with a number if that's easier — 1, 2 or 3." —— 把解码交回去。
"No news is genuinely no news." —— 缺省不是信号。
· 一级 · 一键退出。「这件事今年是不是先不推了？是的话回一句『先不推』，我立刻收手，不用解释——这很合理。」把「拒绝你」变成「帮你省事」，并明说他有权拒绝。
· 二级 · 回一个数字。「回个数字：1 排不上号 / 2 内部卡住 / 3 方向变了。」选项要包含对你最不利的那个——少了它就是诱导题，人一眼看得出来。
· 三级 · 带东西且免回复。「你们那行新出的规我做了一页对照，附件，不用回。」它不为拿回复，只为维持「打开有收获」的记录。
· 四级 · 直接问状态。「这件事到哪一步了？」最贵的一句，留到拿到任何信号之后再用。沉默期直接上四级，等于向欠你钱的人再借一笔。
「王总在吗」→「方便聊两句吗」→「不知道您那边还有没有在推进？」三条消息，三次都要求他写一段话，每一条都在提醒他：你欠我一个回复。欠债的人会躲，这是人之常情。你每发一条，债更重一分，躲的成本更低一分。
「王总，两个月没打扰你了。这件事今年是不是先不推了？是的话回一句『先不推』我就收手，不用解释——这很合理。要是只是排不上号，我年底预算那阵再敲门。」
—— 他回「先不推」，你省下半年瞎猜；他回「不是不推，是内部卡住了」，你拿回了单子。两种都比继续猜好——这正是它管用的原因：你真心接受两种。
EN People don't stay silent because they don't want to reply — they stay silent because replying costs more right now than it returns: recall the thread, ask someone internally, take a position, find a polite tone. You can't remove those costs from their side, but you can rewrite the message so a reply costs one keystroke. Four tiers, cheapest first: a one-word exit ("should I take this off the list for this year?"), a numbered menu (1 not a priority, 2 stuck internally, 3 dead), something useful with the reply obligation waived, and only last the expensive open question "where is this now?". Two peer-reviewed findings sit underneath. Explicitly affirming their freedom to refuse reliably raises compliance across 42 studies; and giving people a genuine choice raises engagement rather than lowering it. The catch is that the exit must be real: if they say "not this year" and you keep sending, the sentence never works on anyone again — because what makes it work is that you meant it.
机制两层。一是成本侧：回复率不由「他多想帮你」决定，由「回这一下要花多少力气」决定，你只能动后者。二是自主感：明说「你完全可以拒绝」会解除心理逆反——被推着走的人本能反向站，被赋予退出权的人反而愿意留下来说一句。所以最容易得到回复的话，恰恰是给对方一个体面走人的机会。
一键退出必须是真退出。他回「先不推」你还继续发，这句话就永远失效；你毁的也不止这一单，是所有话的可信度。另一条边界：这四级压的是回复成本，不是决策成本——他回「2 内部卡住」之后真正的活才开始（找到决策链和挖痛点那两期解的就是这步），破冰只是把灯打开。
"Should I take this off my list for this year?" —— 一键退出的标准问法。
"A one-word yes is completely fine — and a reasonable call." —— 把拒绝变成帮忙。
"Of course, you're free to say no." —— 明确交出拒绝权。
· 第一行：点名触发点。「看到你们上个月把华东那块并进来了——」而不是「好久没联系了——」。前者是他世界里发生的事，后者是你世界里的，他只对前者有反应。
· 第二行：回到他的原话。「当时你说手工对账在两个人手里还扛得住。」证明你记住的是他的问题，不是你的产品。分量全在准确：记错比不记更糟。
· 第三行：低成本问句 + 一个退出。「现在还扛得住吗？不相关的话说一句，我就把它从提醒里删掉。」
EN Your dead deals mostly didn't lose to a competitor — they lost to doing nothing. Inaction is the default option, and switching costs decision effort, migration pain, and the risk of being the person who signed off on a mistake. So the lever for reviving a dead deal is never your enthusiasm; it's what changed on their side. Triage them into four kinds first, because each has a different switch: they chose the status quo (revive when the status quo starts hurting), they had no budget (revive at a new budget cycle), you never had access to the decision maker (revive when people change), or they went with someone else (revive when the honeymoon ends, six to twelve months in, or ninety days before renewal). No trigger, no knock — set a reminder instead. Knocking with nothing new just gets you the same no, harder.
机制：不动是有引力的默认项。改变要付决策成本、迁移成本，还要担「万一是我拍板搞砸的」这层责任；不动出了事，责任是弥散的。再叠上损失比等量收益更重，天平天然偏向不动。真正推得动它的不是你的说服强度，是他那边的外部变化。
还有不作为偏差（Ritov & Baron, 1990）：同样的坏结果，「做了什么」造成的比「什么都没做」造成的更难接受——买错了要被点名，没买只是没发生。「触发点」指他那边可核实的客观变化：换人、并购、融资、上新系统、出事故、预算周期到点。「距离上次联系三个月了」不是触发点，那是你的日历在响。更狠的自查：一个单子三次都停在同一人身上、你从没见过掏钱的那位，它不是死单，是从没活过。
"Saw you merged the east-region team last month —" —— 用他那边的变化开场。
"You said two people could still handle it manually. Is that still true?" —— 回到他的原话。
"If it's not relevant, say so and I'll drop it." —— 三行里的那个退出口。
「王总，这件事我这边不再跟进了，也不用回我解释。如果愿意帮个小忙：回一个字母。A 时机不对 / B 选了别家 / C 内部没通过 / D 我们没讲清楚。这对我改进比成一单还有用。那份口径表随时还是你的，不附带条件。」
· 为什么在放手之后才问。你还想成交时，对方给的是社交性答案——「太贵了」是最省事的拒绝理由，通常不是真的。你说完「不追了」，他不用再保护你的情绪，也不用保护谈判位置。这是经验之谈，没有实验支撑。
· 别问「你为什么不选我们」。那是请他现场编个理由。要问他实际做了什么：「你们最后上了什么？还是先维持原样？」行为比理由可靠得多——这正是 The Mom Test 那期的核心。
· 只要字母，不要长篇。回复成本一高就什么都拿不到。字母攒够十行，你会看见一个自己一直不肯承认的模式。
EN The post-mortem you write for yourself is not trustworthy. Self-serving attribution is one of the most robust findings in social psychology — a meta-analysis of 266 studies confirms people credit wins to themselves and losses to circumstances. Notice what "too expensive" and "bad timing" have in common: neither requires you to change anything, which is exactly why you keep picking them, and keep losing the same deal. There's a second, purely statistical problem: you only learn from deals that closed, so your playbook is copied off the planes that made it back. The fix is cheap. When you close a deal out, waive the explanation and ask for a single letter — A timing, B chose someone else, C no internal sign-off, D we didn't make the case. Ask after you've genuinely let go, because until then they're protecting both your feelings and their negotiating position. And ask what they actually did instead, never why they didn't pick you.
机制两层。一是归因：你写的败因会被自利偏差推向外部原因，于是复盘越勤、错得越稳。二是抽样：你只从成交的单里总结经验，样本被截断在「活下来的」那一侧，规律必然偏。两个漏洞都补不了，只能从外面拿一个低成本的真实分类。
拿到字母就要真的动：C 多了改决策链打法，B 多了改差异化与报价，D 多了改你的讲法，A 多了——最难承认的一种——改资格判定，你一直在追时机不对的人。还有：归档信里只要漏出一丝「看来你们不太重视」的味道，你就把「只是排不上号」的人永久改判成了不。收尾要干净，不带情绪也不带暗示。
"I'm closing this out on my side — no reply needed." —— 干净收尾。
"If you're willing: just a letter. A timing, B someone else, C no sign-off, D we didn't make the case." —— 一键败因。
"What did you end up doing instead?" —— 问行为，不问理由。
打开对话列表，找出所有超过三周没回音的人。20 分钟，纸上完成。