关于「峰值与终点:二次购买记住了什么」,《精品店美学、触觉想象与转化心理》给出的定量结论是什么?

结论:第六章用峰终定律说明,复购翻倍需要每点记忆值 0.1918 的斜率,而小样本有 0.3347 的机会出现假的翻倍。

数据口径:本卷写给在线上把精品包袋卖给台湾与东南亚客户的精品店与代购,研究三件事:页面的留白、字体与颜色怎样表明这是精品店而不是转售商;照片怎样让客户模拟触摸她摸不到的东西;刻意的摩擦与令人难忘的开箱,是否降低后悔并让客户回来。 书中的客户、价格、比率与参数均为示意模型;全部计算由书中程序生成;本书不是法律、财务或心理建议。

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: 0.1918

instant cart peak-end4.5time mean -1.854
concierge peak-end 9.0 time mean 1.224
slope binstantconciergeratio
0.000.16140.16741.037
0.100.16140.23971.485
0.150.16140.28311.754
0.200.16140.33092.050
0.300.16140.43682.706
slope needed to double the rate: 0.1918 per point
查看完整验算程序
# Chapter 6. p11: Remembered experience and the second purchase. Each
# journey is a list of moments (utility from -10 to 10, length in
# days). By the peak-end rule the remembered value is the mean of the
# best moment and the last one; by duration-weighted averaging it is
# the time mean. The 180-day repurchase rate is logistic in the
# remembered value with slope b per point, anchored at 0.1614 for the
# instant cart and starting from 0.1674 for the concierge (selection,
# p10). Utilities and slopes are illustrative. No sampling.
import math

INSTANT = (("click and pay", 5, 0.1), ("silent wait", -2, 12.0),
           ("courier box", 3, 0.1), ("plain unboxing", 4, 0.1))
CONCIERGE = (("personal reply", 6, 0.2), ("deposit and hold slip", 5, 0.1),
             ("store receipt photos", 7, 0.1), ("wait with updates", 1, 12.0),
             ("scented unboxing", 9, 0.1))

def peak_end(j):
    return (max(u for _, u, _ in j) + j[-1][1]) / 2

def time_mean(j):
    return math.fsum(u * t for _, u, t in j) / math.fsum(t for _, _, t in j)

def logit(p):
    return math.log(p / (1 - p))

def inv(x):
    return 1 / (1 + math.exp(-x))

if __name__ == "__main__":
    for name, j in (("instant cart", INSTANT), ("concierge", CONCIERGE)):
        print("   {:12s} peak-end {:4.1f}   time mean {:6.3f}".format(
            name, peak_end(j), time_mean(j)))
    gap = peak_end(CONCIERGE) - peak_end(INSTANT)
    print("     slope b   instant   concierge   ratio")
    for b in (0.0, 0.10, 0.15, 0.20, 0.30):
        r0 = 0.1614
        r1 = inv(logit(0.1674) + b * gap)
        print("     {:7.2f}   {:7.4f}   {:9.4f}   {:5.3f}".format(
            b, r0, r1, r1 / r0))
    need = (logit(2 * 0.1614) - logit(0.1674)) / gap
    print("   slope needed to double the rate: {:.4f} per point".format(need))

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《女性精品买手店计算美学、触觉心理模拟与稀缺转化心理学》第 6 章「峰值与终点:二次购买记住了什么」(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-111 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026111,
  author    = {Fan, Yafu},
  title     = {Computational Aesthetics of Boutique Luxury Web Design, Haptic Mental Imagery and Female Conversion Psychology},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论