关于「高客单破局与八种退出事故」,《断食节律、矿物质与股权止损》给出的定量结论是什么?

结论:第七章算出握有 60.5% 收入的合伙人离开、半数客户跟他走时,期望损失 RM 552,000。

数据口径:「处方库」指可以照做、事后能逐项核对的清单,不是医疗或法律意见。 书中以标注框区分已发表的证据、可运行的计算、原典与作者的假说,每条假说都写明什么结果会推翻它。

English version

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: 552000, 60.5

clients 14; revenue RM 1,824,000; largest 23.0%; top three 51.3%
HHI 1,230; equivalent equal clients 8.1; 14 equal clients would give HHI 714
partner L (delivery) holds 6 accounts, RM612,000 = 33.6%
partner Q (sales) holds 6 accounts, RM 1,104,00060.5%
partner S (finance) holds 2 accounts, RM108,000 = 5.9%
p0.2: expected loss RM 220,800 (12.1% of revenue); P(loss >= 25%) = 0.155
p0.5: expected loss RM 552,000 (30.3% of revenue); P(loss >= 25%) = 0.625
p0.7: expected loss RM 772,800 (42.4% of revenue); P(loss >= 25%) = 0.872
if only C01 and C02 leave: RM 720,00039.5% of revenue
查看完整验算程序
# Chapter 7. p71: How concentrated a high-ticket client book is, and what one departing partner can take.
# Fictional client book of a three-partner firm: annual revenue in RM and the partner who holds the
# relationship. Concentration uses the Herfindahl-Hirschman index on revenue shares (sum of squared
# percentage shares; 10,000 = one client), the same arithmetic the 2010 US merger guidelines apply to
# markets (below 1,500 unconcentrated, above 2,500 highly concentrated); 10,000 / HHI is the number of
# equal-sized clients that would give the same concentration.
# Exit scenario: partner Q leaves; each account Q holds follows Q independently with probability p.
# The loss distribution is enumerated exactly over all subsets of Q's accounts.
from itertools import combinations

BOOK = [("C01", 420000, "Q"), ("C02", 300000, "Q"), ("C03", 216000, "L"), ("C04", 180000, "Q"),
        ("C05", 144000, "L"), ("C06", 120000, "Q"), ("C07", 96000, "L"), ("C08", 72000, "S"),
        ("C09", 72000, "L"), ("C10", 60000, "Q"), ("C11", 48000, "L"), ("C12", 36000, "L"),
        ("C13", 36000, "S"), ("C14", 24000, "Q")]
OWNERS = {"L": "partner L (delivery)", "Q": "partner Q (sales)", "S": "partner S (finance)"}


def loss_distribution(accts, p):
    out = []
    for k in range(len(accts) + 1):
        for sub in combinations(accts, k):
            pr = p ** k * (1 - p) ** (len(accts) - k)
            out.append((sum(sub), pr))
    return out


if __name__ == "__main__":
    total = sum(r for _, r, _ in BOOK)
    shares = sorted((r / total for _, r, _ in BOOK), reverse=True)
    hhi = sum((100 * s) ** 2 for s in shares)
    print("   clients {}; revenue RM {:,}; largest {:.1%}; top three {:.1%}".format(
        len(BOOK), total, shares[0], sum(shares[:3])))
    print("   HHI {:,.0f}; equivalent equal clients {:.1f}; 14 equal clients would give HHI {:,.0f}".format(
        hhi, 10000 / hhi, 10000 / 14))
    for o, name in OWNERS.items():
        rev = sum(r for _, r, w in BOOK if w == o)
        print("   {:22s} holds {} accounts, RM {:>9,} = {:.1%}".format(
            name, sum(1 for *_, w in BOOK if w == o), rev, rev / total))
    q = [r for _, r, w in BOOK if w == "Q"]
    for p in (0.2, 0.5, 0.7):
        dist = loss_distribution(q, p)
        mean = sum(v * pr for v, pr in dist)
        big = sum(pr for v, pr in dist if v >= 0.25 * total)
        print("   p = {}: expected loss RM {:,.0f} ({:.1%} of revenue); P(loss >= 25%) = {:.3f}".format(
            p, mean, mean / total, big))
    print("   if only C01 and C02 leave: RM {:,} = {:.1%} of revenue".format(720000, 720000 / total))

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《断食代谢节律、靶向矿物质与高客单商业破局母典》第 7 章「高客单破局与八种退出事故」(版本 0.1.0). 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-130 · v0.1.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026130,
  author    = {Fan, Yafu},
  title     = {The Fasting-Rhythm, Targeted-Mineral and High-Ticket Business Override Codex},
  year      = {2026},
  publisher = {5D Library},
  edition   = {0.1.0},
  note      = {DOI pending}
}

本书其他章节结论