关于「信息缺口的剂量:只用访客自己的免费结果写悬念召回」,《弃单召回与升单棘轮》给出的定量结论是什么?

结论:\(\gamma\) 取 1.5 时,拉力在 \(g^{*}=1/(1+\gamma)=0.400\) 处最大。表中,剂量 0 时点击率只有基础值 0.0200,净值 +0.0080;剂量 0.2 时点击率升到 0.0893;剂量 0.4 时达到顶点 0.1100,投诉 0.00059,净值 +0.1083;剂量 0.6 时点击率回落到 0.0935,投诉升到 0.00105;剂量 0.8 时点击率只剩 0.0546,投诉 0.00194,净值转负为 -0.0119;剂量 1.0 时点击率又回到 0.0200,投诉 0.00340,净值 -0.1120。

数据口径:按本书的示意模型,第一波召回放在一分钟时只挽回同意离开者的 0.00609,放在十五分钟时挽回 0.03732;把三波压在一小时之内,每位离开者净值 RM 1.1790,按十五分钟、二十四小时、七十二小时发送时为 RM 1.9419;第五波的净值已是 -0.0121。 书中网站被当作付费报告产品讨论,不宣称任何预测准确性,不用恐惧招揽;所有访客与订单数据均为合成或示意。

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。 与结论吻合的打印值: -0.0119, -0.1120, 0.00059, 0.00105, 0.00194, 0.00340

peak of pull at g* = 1 / (1 + gamma) = 0.400 (gamma = 1.5)
dose gclickcomplaintnet RM per message
0.00.02000.00040+0.0080
0.10.06130.00040+0.0575
0.20.08930.00042+0.0902
0.30.10510.00048+0.1068
0.40.11000.00059+0.1083
0.50.10560.00078+0.0957
0.60.09350.00105+0.0703
0.70.07570.00143+0.0337
0.80.05460.00194-0.0119
0.90.03380.00259-0.0629
1.00.02000.00340-0.1120
net value peaks at g = 0.36 (net RM 0.1095), left of the click peak
five named facts: open questions 0..5 give doses 0.0, 0.2, 0.4, 0.6, 0.8, 1.0
gamma 0.5: click peak 0.667, net peak 0.51
gamma 1.0: click peak 0.500, net peak 0.42
gamma 3.0: click peak 0.250, net peak 0.24
查看完整验算程序
# Chapter 3. p31: an inverted-U model of the information-gap dose.
# A recall message names m facts from the visitor's own free result; for
# some of them it states an open question that the full report answers.
# Dose g = open questions / named facts, between 0 and 1.
# Illustrative model (synthetic parameters, after Loewenstein 1994 and the
# inverted U in confidence reported by Kang et al. 2009):
#   pull(g) = g * (1 - g)^gamma: no gap, no pull; a gap with nothing
#     already known around it gives nothing to hold on to.
#   click(g) = 0.020 + 0.090 * pull(g) / max pull.
#   a vague message annoys: complaint(g) = 0.0004 + 0.0030 * g^3.
#   value per click RM 1.20 (later purchases); per complaint RM 40.00.
GAMMA = 1.5
G_PEAK = 1.0 / (1.0 + GAMMA)


def pull(g, gamma=GAMMA):
    return g * (1.0 - g) ** gamma


def click(g, gamma=GAMMA):
    top = pull(1.0 / (1.0 + gamma), gamma)
    return 0.020 + 0.090 * pull(g, gamma) / top


def complaint(g):
    return 0.0004 + 0.0030 * g ** 3


def value(g, gamma=GAMMA):
    return 1.20 * click(g, gamma) - 40.0 * complaint(g)


if __name__ == "__main__":
    print("peak of pull at g* = 1 / (1 + gamma) = {:.3f} (gamma = {})".format(G_PEAK, GAMMA))
    print("dose g   click    complaint   net RM per message")
    for i in range(11):
        g = i / 10.0
        print(" {:.1f}     {:.4f}   {:.5f}     {:+.4f}".format(g, click(g), complaint(g), value(g)))
    best = max((value(i / 100.0), i / 100.0) for i in range(101))
    print("net value peaks at g = {:.2f} (net RM {:.4f}), left of the click peak".format(best[1], best[0]))
    print("five named facts: open questions 0..5 give doses 0.0, 0.2, 0.4, 0.6, 0.8, 1.0")
    for gm in (0.5, 1.0, 3.0):
        bestg = max((value(i / 100.0, gm), i / 100.0) for i in range(101))[1]
        print("   gamma {:.1f}: click peak {:.3f}, net peak {:.2f}".format(gm, 1.0 / (1.0 + gm), bestg))

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《诊断收费站弃单挽回时序、个性化悬念召回与高客单升单棘轮母典》第 3 章「信息缺口的剂量:只用访客自己的免费结果写悬念召回」(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-349 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026349,
  author    = {Fan, Yafu},
  title     = {Paywall Abandonment Recovery Timing, Personalised Curiosity Recall and High-Ticket Upsell Ratchets},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论