关于「均匀撒钱为什么输:巴斯扩散上的同额对照」,《新品冷启动的广告脉冲与口碑接棒》给出的定量结论是什么?
结论:巴斯方程里,买家从自发尝鲜与口碑两个方向来,而口碑是引擎。广告线性作用于创新项时,同一笔钱越早花,乘数越大;前六天打满比六十天均匀多赚 13.4 %,放在最后六天几乎白花。可是「前六天」这个数从哪里来?为什么是六天而不是五天或八天,第六天之后为什么一分都不该再花?要回答它,需要一个能逐日给每一元钱定价的工具,这就是下一章的哈密顿量与开关函数。
数据口径:本书用巴斯扩散加广告控制的最优控制模型检验这一做法:广告以加法、线性方式作用于创新项并有每日上限时,庞特里亚金极大值原理与贝尔曼动态规划给出同一个答案,开头打满上限、在开关函数变负的那天一次停投,把接力棒交给口碑;示意参数下同额预算前六天打满比六十天均匀多赚 13.4 %。 书中的市场、毛利、扩散系数、广告效果、上限与产能均为示意参数;全部计算由书中程序生成;本书是非个人化经营建议,紧迫感只来自真实库存与真实截止。
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: 13.4
schedule profit reached vs even front pulse 6 days 531909 0.9365 +13.4 % even 500 per day 468874 0.8315 +0.0 % middle pulse 27-32 422081 0.7535 -10.0 % late pulse 54-59 352496 0.6375 -24.8 % no advertising 349990 0.5833 -25.4 % best even (950 per day) 482598 0.8993 +2.9 % budget front-pulse gain over even spread of the same budget 10000 +21.1 % 20000 +17.1 % 30000 +13.4 % 40000 +10.8 % 60000 +7.3 %
核心公式
\[\frac{dx}{dt}=(p+qx)(1-x).\]
查看完整验算程序
# Chapter 1. p12: Same budget, different timing: pulse, even and late.
# Model: daily Bass diffusion with advertising in the innovation term,
# x[t+1] = x[t] + (p + beta u[t] + q x[t]) (1 - x[t]),
# profit = margin * M * (x[t+1] - x[t]) - u[t], summed over a 60-day
# window with no salvage value. All parameters are illustrative:
# M = 20000 buyers, margin 30 yuan, p = 0.001, q = 0.08 per day,
# beta = 4e-6 per yuan, a daily cap of 5000 yuan. Every schedule below
# spends the same 30000 yuan, except the zero and best-even baselines.
import math
M, MARGIN, P, Q, BETA, UMAX, T = 20000, 30.0, 0.001, 0.08, 4e-6, 5000.0, 60
def run(us):
x, cash = 0.0, []
for u in us:
x2 = x + (P + BETA * u + Q * x) * (1 - x)
cash.append(MARGIN * M * (x2 - x) - u)
x = x2
return math.fsum(cash), x
B = 30000.0
k = int(B / UMAX)
plans = [
("front pulse 6 days", [UMAX] * k + [0.0] * (T - k)),
("even 500 per day", [B / T] * T),
("middle pulse 27-32", [0.0] * 27 + [UMAX] * k + [0.0] * (T - 27 - k)),
("late pulse 54-59", [0.0] * (T - k) + [UMAX] * k),
("no advertising", [0.0] * T),
]
best_c = max(range(0, 5001, 10), key=lambda c: run([float(c)] * T)[0])
plans.append(("best even (%d per day)" % best_c, [float(best_c)] * T))
base = run(plans[1][1])[0]
print("schedule profit reached vs even")
for name, us in plans:
v, x = run(us)
print("%-24s %9.0f %.4f %+.1f %%" % (name, v, x,
100 * (v / base - 1)))
print("budget front-pulse gain over even spread of the same budget")
for b in (10000, 20000, 30000, 40000, 60000):
n = int(b / UMAX)
f = run([UMAX] * n + [0.0] * (T - n))[0]
e = run([b / T] * T)[0]
print("%6d %+.1f %%" % (b, 100 * (f / e - 1)))唯一数据源与引用
范雅福(Fan, Y.). (2026). 《庞特里亚金极大值原理、新品冷启动广告脉冲烧钱速率与口碑接棒最优控制母典》第 1 章「均匀撒钱为什么输:巴斯扩散上的同额对照」(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-241 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026241,
author = {Fan, Yafu},
title = {Pontryagin's Maximum Principle at Product Launch: Optimal Ad-Pulse Burn Rates and the Hand-Over to Word of Mouth},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}