关于「满减阶梯毛利配比:让利漏给了谁」,《熟客唤醒与凑单闪购》给出的定量结论是什么?
结论:满减阶梯的让利,大半漏给了本来就在门槛上面的人。示意参数下,一张「满 150 减 10、满 250 减 25、满 400 减 50」的体面阶梯,让营收略增、毛利明显下降;每换回一块钱凑单毛利,第一档要让出 4.27 块,第三档要让出 21.90 块。按营收设计的阶梯几乎一定亏毛利;按毛利设计的阶梯让利很小、档位很高,而且只在凑单区有真正尾货时才值得设。
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: 150, 21.90, 250, 4.27, 400
tier share saving/visitor filler margin net below 150 0.7678 0.0000 0.0000 +0.0000 150 / -10 0.1905 1.9049 0.4461 -1.4588 250 / -25 0.0379 0.9469 0.0948 -0.8520 400 / -50 0.0038 0.1909 0.0087 -0.1822 saving rate per tier: 0.067, 0.100, 0.125 saving given per RM of filler margin: 4.27, 9.98, 21.90 no ladder: margin 29.664 revenue 98.88 per visitor ladder : margin 27.171 revenue 99.50 per visitor change : margin -2.493 revenue +0.62 filler margin 0.30: margin change -1.943 filler margin 0.45: margin change -1.394
查看完整验算程序
# Chapter 5. p51: a spend-and-save ladder, tier by tier, in margin terms.
# Ladder (illustrative): spend RM 150 save RM 10, RM 250 save RM 25,
# RM 400 save RM 50. Intended basket lognormal (median RM 85, log-sd
# 0.55), item margin 0.30, filler margin 0.15, top-up to the next tier
# with chance (1 - exp(-saving / PSI)) * exp(-gap / 25), PSI = RM 8 (the
# pull of the saving saturates), filler = gap + 4. Shipping is left out
# (chapter 4 handles it). A buyer already past a tier gets its saving
# whether or not the ladder moved her: that is the leak. Per tier we
# report: share of visitors ending there, leak, filler margin, net.
import math
MED, SD, MU, MUF, DELTA, OVER = 85.0, 0.55, 0.30, 0.15, 25.0, 4.0
PSI = 8.0
LADDER = [(150.0, 10.0), (250.0, 25.0), (400.0, 50.0)]
Z = [i / 200.0 for i in range(-1200, 1201)]
WZ = [math.exp(-z * z / 2.0) / math.sqrt(2.0 * math.pi) / 200.0 for z in Z]
BASKETS = [MED * math.exp(SD * z) for z in Z]
def tier_of(b, ladder):
k = -1
for j, (t, _) in enumerate(ladder):
if b >= t:
k = j
return k
def evaluate(ladder, mu=MU, muf=MUF):
nt = len(ladder)
share = [[] for _ in range(nt + 1)]
leak = [[] for _ in range(nt + 1)]
fill = [[] for _ in range(nt + 1)]
marg, rev = [], []
for w, b in zip(WZ, BASKETS):
k = tier_of(b, ladder)
d_now = ladder[k][1] if k >= 0 else 0.0
if k + 1 < nt:
t_next, d_next = ladder[k + 1]
gap = t_next - b
tau = (1.0 - math.exp(-d_next / PSI)) * \
math.exp(-gap / DELTA)
else:
tau, gap, d_next = 0.0, 0.0, 0.0
if k + 1 < nt: # topped up: lands on tier k + 1
share[k + 2].append(w * tau)
fill[k + 2].append(w * tau * muf * (gap + OVER))
leak[k + 2].append(w * tau * d_next)
share[k + 1].append(w * (1.0 - tau))
leak[k + 1].append(w * (1.0 - tau) * d_now)
marg.append(w * (mu * b + tau * (muf * (gap + OVER) - d_next)
- (1.0 - tau) * d_now))
rev.append(w * (b + tau * (gap + OVER - d_next)
- (1.0 - tau) * d_now))
return ([math.fsum(x) for x in share], [math.fsum(x) for x in leak],
[math.fsum(x) for x in fill], math.fsum(marg), math.fsum(rev))
if __name__ == "__main__":
base_m = math.fsum(w * MU * b for w, b in zip(WZ, BASKETS))
base_r = math.fsum(w * b for w, b in zip(WZ, BASKETS))
share, leak, fill, m, r = evaluate(LADDER)
print("tier share saving/visitor filler margin net")
names = ["below 150", "150 / -10", "250 / -25", "400 / -50"]
for j in range(4):
print("{:<12} {:.4f} {:.4f} {:.4f} {:+.4f}"
.format(names[j], share[j], leak[j], fill[j],
fill[j] - leak[j]))
print("saving rate per tier: " + ", ".join(
"{:.3f}".format(d / t) for t, d in LADDER))
print("saving given per RM of filler margin: " + ", ".join(
"{:.2f}".format(leak[j] / fill[j]) for j in range(1, 4)))
print("no ladder: margin {:.3f} revenue {:.2f} per visitor".format(
base_m, base_r))
print("ladder : margin {:.3f} revenue {:.2f} per visitor".format(m, r))
print("change : margin {:+.3f} revenue {:+.2f}".format(
m - base_m, r - base_r))
for muf in (0.30, 0.45):
_, _, _, m2, _ = evaluate(LADDER, muf=muf)
base2 = base_m
print("filler margin {:.2f}: margin change {:+.3f}".format(
muf, m2 - base2))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 5 章「满减阶梯毛利配比:让利漏给了谁」(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}
}