关于「尾单清仓:有限期定价、两小时脉冲与等待熟客」,《熟客唤醒与凑单闪购》给出的定量结论是什么?
结论:尾单清仓的价格路径取决于剩余件数与剩余时间:件数越少、时间越多,价格越高;库存相对客流充足时,固定价格已经接近最优,最后时刻也不该甩到地板价。两小时脉冲在示意参数下卖出 9.21 件,半小时只卖出 4.79 件,窗口太短,看见的人太少。闪购太频繁,熟客学会等待,正价复购被侵蚀;学得较快时,每月只宜一场。
数据口径:示意参数下,面额 RM 6 的券只在间隔比 0.85 到 11.37 之间净值为正,刚买客与久睡客都是负值;不设对照组的核销记账会把增量高估 12.2 倍。 书中参数为示意,不对应 HOMIE-X 任何真实客户、订单或供货商,书中价格一律为示意;全部计算由书中程序生成。
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: 4.79, 9.21
| price RM 60 | ||||
| hours | demand mean | P(sell out) | units sold | revenue RM |
| 0.5 | 4.80 | 0.025 | 4.79 | 328.93 |
| 1.0 | 7.82 | 0.261 | 7.44 | 466.99 |
| 2.0 | 11.20 | 0.681 | 9.21 | 559.11 |
| 4.0 | 13.76 | 0.879 | 9.75 | 586.82 |
| 8.0 | 14.65 | 0.918 | 9.84 | 591.49 |
| 2-hour window, by price | ||||
| price | demand mean | P(sell out) | units sold | revenue RM |
| 40 | 15.63 | 0.948 | 9.90 | 396.85 |
| 50 | 13.23 | 0.849 | 9.68 | 486.39 |
| 60 | 11.20 | 0.681 | 9.21 | 559.11 |
| 70 | 9.48 | 0.476 | 8.50 | 606.70 |
| 80 | 8.03 | 0.287 | 7.59 | 626.74 |
| revenue-best 2-hour price RM 85 (627.55) |
查看完整验算程序
# Chapter 6. p62: a pulse flash sale of a real tail lot, by window length.
# A lot of 10 units, announced to a consenting group. Viewers in a
# window of h hours: Poisson with mean V(h) = 40 (1 - exp(-h / 1.5))
# (attention saturates). A viewer buys one unit at price p with chance
# exp(-p / 60) * (1 + 0.25 exp(-h)): a short, true deadline adds at most
# a quarter to conversion. Units sold = min(10, Poisson demand); unsold
# units go to the clearance channel at RM 8. Exact Poisson sums, no
# simulation. All parameters illustrative.
import math
LOT, VMAX, TAU, THETA, SALV = 10, 40.0, 1.5, 60.0, 8.0
def pois_pmf(k, mu):
return math.exp(-mu + k * math.log(mu) - math.lgamma(k + 1.0))
def flash(h, p, lot=LOT):
mu = VMAX * (1.0 - math.exp(-h / TAU)) * math.exp(-p / THETA) \
* (1.0 + 0.25 * math.exp(-h))
below = [pois_pmf(k, mu) for k in range(lot)]
p_out = 1.0 - math.fsum(below)
sold = math.fsum([k * below[k] for k in range(lot)]) + lot * p_out
rev = p * sold + SALV * (lot - sold)
return mu, p_out, sold, rev
if __name__ == "__main__":
print("price RM 60")
print("hours demand mean P(sell out) units sold revenue RM")
for h in (0.5, 1.0, 2.0, 4.0, 8.0):
mu, po, s, r = flash(h, 60.0)
print("{:>5} {:6.2f} {:.3f} {:5.2f} {:7.2f}"
.format(h, mu, po, s, r))
print("2-hour window, by price")
print("price demand mean P(sell out) units sold revenue RM")
for p in (40.0, 50.0, 60.0, 70.0, 80.0):
mu, po, s, r = flash(2.0, p)
print("{:>5.0f} {:6.2f} {:.3f} {:5.2f} {:7.2f}"
.format(p, mu, po, s, r))
best = max(range(20, 121, 5), key=lambda p: flash(2.0, float(p))[3])
print("revenue-best 2-hour price RM {} ({:.2f})".format(
best, flash(2.0, float(best))[3]))唯一数据源与引用
范雅福(Fan, Y.). (2026). 《私域代购熟客复购唤醒、满额凑单毛利配比与限时清仓脉冲闪购母典》第 6 章「尾单清仓:有限期定价、两小时脉冲与等待熟客」(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-353 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026353,
author = {Fan, Yafu},
title = {Repeat-Buyer Reactivation in Private-Channel Commerce, Threshold Bundle Margin Design and Flash-Clearance Dynamics},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}