关于「预览做到几成:完成度饱和曲线、最优完成度与预览合规」,《演示先行与四十八小时交付》给出的定量结论是什么?

结论:预览的赢率沿饱和曲线上升,工时沿凸曲线上升,期望净值在中间达到最大。示意参数下最优完成度约为 0.289,做满反而比不做还亏。毛利越高,最优完成度越高;赢率越快饱和,越应早停。宜家效应要求客户自己出力且劳动完成,不能直接借来论证半成品预览,只能写成「客户参与」的假说去检验。预览给客户看之前,要过素材来源、商标、索引、对比度与点按高度五道检查。

数据口径:书中的赢率、工时、毛利、首付比例、月费与价格均为示意参数,以「单位」计,不对应任何币种、真实客户或真实成交;全部计算由书中程序生成。 本书不是法律意见,也不是个人化经营或理财建议,合同条款须由当地执业律师审阅;书中不写威胁催促、不写会重置的倒计时、不写收益承诺,推荐奖励只写单层。

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: 0.289

k4, margin 4,400: x* = 0.289 (grid 0.289)
p(x*)0.326, hours 4.89, E = 1,139.3
no preview E820.0; full preview E = 292.0
gain of x* over no preview319.3 units
x* by margin (rows) and curvature k (columns)
margin k2 k=4 k=8
2,0000.078 0.148 0.156
4,4000.292 0.289 0.238
8,0000.476 0.401 0.301
catalogue hours 15.5; shown working 4.5 -> completeness 0.290
查看完整验算程序
# Chapter 2. p21: how complete should the preview be?
# Win rate rises with completeness x in [0, 1] along a saturating curve
#   p(x) = p0 + (p1 - p0) * (1 - exp(-k x)) / (1 - exp(-k))
# and preview hours rise convexly: h(x) = 1 + 10 x + 12 x^2.
# Expected net value E(x) = p(x) * M - RATE * h(x). All parameters are
# illustrative (p0 0.20 without preview, p1 0.38 at a full preview).
# The optimum is found on a grid of 0.001 and refined by bisection on
# the derivative; a sensitivity table varies margin M and curvature k.
import math

P0, P1, RATE = 0.20, 0.38, 60


def p_win(x, k):
    return P0 + (P1 - P0) * (1 - math.exp(-k * x)) / (1 - math.exp(-k))


def hours(x):
    return math.fsum([1.0, 10.0 * x, 12.0 * x * x])


def value(x, k, m):
    return p_win(x, k) * m - RATE * hours(x)


def deriv(x, k, m):
    dp = (P1 - P0) * k * math.exp(-k * x) / (1 - math.exp(-k))
    return dp * m - RATE * (10.0 + 24.0 * x)


def best(k, m):
    if deriv(0.0, k, m) <= 0:
        return 0.0
    if deriv(1.0, k, m) >= 0:
        return 1.0
    lo, hi = 0.0, 1.0
    for _ in range(60):
        mid = (lo + hi) / 2
        if deriv(mid, k, m) > 0:
            lo = mid
        else:
            hi = mid
    return (lo + hi) / 2


K, M = 4.0, 4400
x = best(K, M)
grid = max((value(i / 1000, K, M), i / 1000) for i in range(1001))
print("k = 4, margin 4,400: x* = {:.3f} (grid {:.3f})".format(x, grid[1]))
print("  p(x*) = {:.3f}, hours {:.2f}, E = {:,.1f}".format(
    p_win(x, K), hours(x), value(x, K, M)))
print("  no preview E = {:.1f}; full preview E = {:.1f}".format(
    value(0.0, K, M), value(1.0, K, M)))
print("  gain of x* over no preview: {:.1f} units".format(
    value(x, K, M) - value(0.0, K, M)))
print("x* by margin (rows) and curvature k (columns)")
print("  margin   k=2     k=4     k=8")
for m in (2000, 4400, 8000):
    cells = ["{:.3f}".format(best(k, m)) for k in (2.0, 4.0, 8.0)]
    print("  {:>6,}  ".format(m) + "   ".join(cells))
# Completeness measured by catalogue hours (modes of the six-module
# order of chapter 3): skin applied 3.0 h shown working, plus a
# read-only slot table worth half of the 3.0 h booking module.
CATALOGUE = [1.0, 3.0, 4.0, 3.0, 2.5, 2.0]
shown = math.fsum([3.0, 1.5])
print("catalogue hours {:.1f}; shown working {:.1f} -> completeness {:.3f}"
      .format(math.fsum(CATALOGUE), shown, shown / math.fsum(CATALOGUE)))

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《半成品演示先行逼单、四十八小时交钥匙系统交付与防需求蔓延母典》第 2 章「预览做到几成:完成度饱和曲线、最优完成度与预览合规」(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-368 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026368,
  author    = {Fan, Yafu},
  title     = {Demo-First B2B Closing, Forty-Eight-Hour Turnkey System Delivery and Scope-Creep Containment},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论