关于「「我再想想」:沉默的后验与选项冲突」,《异议柔道与定金锁单》给出的定量结论是什么?
结论:第三章读解沉默:说完「再想想」后第七天,七天内回来的概率只剩 0.091;方案从两个增加到六个,成交概率从 0.635 降到 0.492。
数据口径:书中的顾客群、概率与金额均为示意;全部计算由书中程序生成;本书不构成个人化的法律、财务或投资建议,只讨论守规矩的说服。 第一章的判别树说明,在示意先验下,说「太贵了」的人有 0.734 并不缺钱。
对照数据表
下表数字由本章 Python 程序原样打印,可重新运行复现。 程序只用 Python 标准库与结论吻合的打印值: 0.492, 0.635
| n | P(buy) | P(defer) | of which conflict |
|---|---|---|---|
| 1 | 0.579 | 0.421 | 0.000 |
| 2 | 0.635 | 0.365 | 0.187 |
| 3 | 0.600 | 0.400 | 0.326 |
| 4 | 0.553 | 0.447 | 0.415 |
| 5 | 0.520 | 0.480 | 0.467 |
| 6 | 0.492 | 0.508 | 0.502 |
| most sales at n = 2; n = 6 keeps 0.774 of that | |||
| six plans, adviser shows two: P(buy) = 0.626 |
查看完整验算程序
# Chapter 3. p32: More options, more "let me think": a conflict model of
# deferral in the spirit of Tversky and Shafir (1992) and Dhar (1997).
# A customer sees n plans whose utilities are independent N(0, 1). She
# buys the best plan if it beats her reservation level U0 and leads the
# runner-up by at least DELTA; otherwise she defers. Parameters are
# illustrative. A needs-led adviser shortlists two of six plans from
# the customer's stated needs, with estimation error.
# Fixed-seed Monte Carlo, 100,000 customers per row.
import random
U0, DELTA, N = -0.20, 0.50, 100000
rng = random.Random(20261112)
print(" n P(buy) P(defer) of which conflict")
rows = {}
for n in range(1, 7):
buy = defer = conflict = 0
for _ in range(N):
u = sorted((rng.gauss(0.0, 1.0) for _ in range(n)), reverse=True)
if u[0] < U0:
defer += 1
elif n > 1 and u[0] - u[1] < DELTA:
defer += 1
conflict += 1
else:
buy += 1
rows[n] = buy / N
print("{:>2} {:>8.3f} {:>9.3f} {:>12.3f}".format(
n, buy / N, defer / N, conflict / N))
best = max(rows, key=lambda k: rows[k])
print("most sales at n = {}; n = 6 keeps {:.3f} of that".format(
best, rows[6] / rows[best]))
# needs-led: the adviser ranks six plans by what the customer said she
# needs, an estimate with error N(0, 0.7), and shows only the top two
buy = 0
for _ in range(N):
u = [rng.gauss(0.0, 1.0) for _ in range(6)]
e = [x + rng.gauss(0.0, 0.7) for x in u]
top = sorted(range(6), key=lambda i: e[i], reverse=True)[:2]
a, b = sorted((u[i] for i in top), reverse=True)
if a >= U0 and a - b >= DELTA:
buy += 1
print("six plans, adviser shows two: P(buy) = {:.3f}".format(buy / N))唯一数据源与引用
范雅福(Fan, Y.). (2026). 《顾问式销售异议柔道、价格嫌贵贝叶斯拆解与定金锁单停时协议》第 3 章「「我再想想」:沉默的后验与选项冲突」(版本 1.0.0). 5D Library. DOI 待发布
DOI 待发布 · 专著页正式发布后上线(本站暂无专著页) · 5DL-B-146 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026146,
author = {Fan, Yafu},
title = {Consultative Objection Judo, a Bayesian Decomposition of "Too Expensive" and Deposit-Lock Stopping-Time Protocols},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}