5D Library › One-Person Enterprise Automation, Avatar Livestream Funnels and Task-Exposure Arbitrage · 中文
Chapter 6. Moving Income Across: Transition Paths and a Solo Portfolio
When to quit
If a job’s tasks are being taken over, the monthly chance of losing it rises. A side business can replace the salary, but it takes cumulative hours to mature, and some offers never find a market. Below, a salary of RM 8,000 a month faces a layoff hazard rising from 0.005 to 0.025 a month over three years; a side business matures along a logistic curve in cumulative hours, and with probability 0.4 it never finds its market. Four strategies are compared over 36 months.
Calculation
# Chapter 6. p11: Three ways through 36 months of rising job risk.
# Salary RM 8,000 a month, living costs RM 5,000, savings RM 30,000. The
# monthly chance of losing the job rises from 0.005 to 0.025 as AI takes
# over tasks; out of work, a new job (at 0.8 of the pay) is found with
# probability 0.15 a month. A side business at maturity earns RM 9,000 a
# month on 40 hours a month, scaled by (hours / 40) ** 0.6, reached along
# a logistic curve in cumulative hours (half at 600 h, scale 150 h), with
# costs of 15% of revenue; with probability 0.4 the
# offer never finds its market and earns only 0.2 of that. Strategies:
# (a) job only; (b) 40 side hours a month, 160 when out of work; (c) quit
# now, 160 a month; (d) as (b), but quit for 160 hours a month once the
# side business has covered living costs (RM 5,000) for 3 months running.
# 20,000 runs each, seed 90 (illustrative parameters).
import math
import random
def side(hours, month_h, fit=1.0):
if hours == 0:
return 0.0
cap = 9000 * (month_h / 40) ** 0.6
return fit * cap / (1 + math.exp(-(hours - 600) / 150)) * 0.85
def run(strategy, rng, months=36):
cash, pay, employed, hours = 30000.0, 8000.0, strategy != "c", 0.0
low, fit = cash, (0.2 if rng.random() < 0.4 else 1.0)
streak, quit = 0, False
for t in range(months):
hazard = 0.005 + 0.02 * t / (months - 1)
if employed and rng.random() < hazard:
employed = False
elif (not employed and strategy != "c" and not quit
and rng.random() < 0.15):
employed, pay = True, pay * 0.8
if strategy == "d" and employed and streak >= 3:
employed, quit = False, True
month_h = 0
if strategy in "bd":
month_h = 40 if employed else 160
elif strategy == "c":
month_h = 160
hours += month_h
income = side(hours, month_h, fit)
streak = streak + 1 if income >= 5000 else 0
cash += (pay if employed else 0) + income - 5000
low = min(low, cash)
return cash, low, income
if __name__ == "__main__":
for s, name in (("a", "job only"), ("b", "job + 40 h side"),
("c", "quit, full time"), ("d", "quit when covered")):
rng = random.Random(90)
rows = sorted(run(s, rng) for _ in range(20000))
mean = sum(r[0] for r in rows) / len(rows)
broke = sum(1 for r in rows if r[1] < 0) / len(rows)
print(" {:16s}: mean cash RM {:7,.0f}, 5th pct RM {:7,.0f}"
.format(name, mean, rows[1000][0]))
print(" {:16s} ran out of cash {:.3f}, side income month 36"
" RM {:,.0f}".format("", broke, sum(r[2] for r in rows)
/ len(rows)))
# job only : mean cash RM 110,540, 5th pct RM 21,200
# ran out of cash 0.037, side income month 36 RM 0
# job + 40 h side : mean cash RM 241,533, 5th pct RM 107,417
# ran out of cash 0.001, side income month 36 RM 6,053
# quit, full time : mean cash RM 241,689, 5th pct RM -34,980
# ran out of cash 0.399, side income month 36 RM 11,970
# quit when covered: mean cash RM 264,410, 5th pct RM 106,174
# ran out of cash 0.001, side income month 36 RM 11,240Keeping only the job ends with RM 110,540 on average, and 0.037 of runs run out of cash after a layoff. Working 40 side hours a month on top of the job ends with RM 241,533 and almost never runs out (0.001). Quitting now for full-time work on the side business ends with almost the same mean, RM 241,689, but runs out of cash in 0.399 of runs: quitting before the market has answered stakes the savings on that answer. The best rule in the model is to quit only after the side business has covered living costs for 3 months running: mean RM 264,410, with the same 0.001 chance of running out as the cautious path. The rule works because it lets the market reveal whether the offer fits before giving up the salary.
A portfolio for one founder
A solo income can mix lumpy high-margin products with steadier client work. Below, the founder has 120 hours a month; a product line takes 30 hours and nets RM 6,000 a month on average with a standard deviation of 4,000; a B2B automation retainer takes 40 hours for RM 5,000 (sd 1,000); content with affiliate income takes 20 hours for RM 1,500. A month below RM 6,000 counts as short.
Calculation
# Chapter 6. p12: A solo portfolio and the months that fall short.
# Founder hours: 120 a month. Units of work: a high-ticket digital
# product (30 h, net RM 6,000 a month, sd 4,000, lumpy launches), a B2B
# automation retainer (40 h, RM 5,000, sd 1,000) and content with
# affiliate income (20 h, RM 1,500, sd 800, correlation 0.5 with the
# product). At most two product lines and two retainer clients are
# assumed manageable. Income below RM 6,000 is a short month. Each
# mix using 101 to 120 hours is simulated for 12 months x 5,000 runs
# (seed 90); mean monthly income and the share of short months.
import itertools
import math
import random
UNITS = (("product", 30, 6000, 4000), ("retainer", 40, 5000, 1000),
("content", 20, 1500, 800))
def simulate(mix, rng, months=60000):
total = short = 0.0
for _ in range(months):
z1, z2, z3 = rng.gauss(0, 1), rng.gauss(0, 1), rng.gauss(0, 1)
zc = 0.5 * z1 + math.sqrt(1 - 0.25) * z3
x = (mix[0] * max(0.0, 6000 + 4000 * z1) +
mix[1] * max(0.0, 5000 + 1000 * z2) +
mix[2] * max(0.0, 1500 + 800 * zc))
total += x
short += x < 6000
return total / months, short / months
if __name__ == "__main__":
rows = []
for mix in itertools.product(range(3), range(3), range(7)):
h = sum(m * u[1] for m, u in zip(mix, UNITS))
if 0 < h <= 120 and h > 120 - 20:
rng = random.Random(90)
rows.append((mix, h) + simulate(mix, rng))
print(" product retainer content hours mean RM short months")
for mix, h, mean, sh in sorted(rows, key=lambda r: -r[2]):
print(" {} {} {} {:3d} {:7,.0f} {:.3f}"
.format(mix[0], mix[1], mix[2], h, mean, sh))
# product retainer content hours mean RM short months
# 2 1 1 120 18,749 0.048
# 2 0 3 120 16,769 0.131
# 1 2 0 110 16,116 0.002
# 1 1 2 110 14,136 0.037
# 0 2 2 120 13,011 0.002
# 1 0 4 110 12,156 0.162
# 0 1 4 120 11,031 0.064
# 0 0 6 120 9,052 0.266Two product lines, one retainer and one content stream give the highest mean (RM 18,749 a month) with 0.048 of months short. One product and two retainers give RM 16,116 with only 0.002 short. Products without retainers are riskier: two products and three content streams average RM 16,769 but fall short in 0.131 of months. Content alone is the worst on both counts (RM 9,052, 0.266). The retainer is what buys safety; the product is what buys upside.
Hypothesis · not established
The author’s interpretation (not yet tested): solo founders who keep at least one recurring B2B client while building a product have fewer months below their living costs in their first two years than founders who build products only, at a similar mean income. What would show it wrong: in monthly income records of solo founders, those with a recurring client have as many short months as those without.
Summary
Quitting before the side business has proved itself gives the same mean as keeping the job and building on the side, with a 0.399 chance of running out of cash. Quitting once side income has covered living costs for 3 months gives the best mean (RM 264,410) at a 0.001 risk. In a solo portfolio, a retainer cuts short months from 0.131 to 0.048 or less.
Fan, Y. (2026). One-Person Enterprise Automation, Avatar Livestream Funnels and Task-Exposure Arbitrage: Coasean Transaction-Cost Collapse, Cyclic Script Retention and High-Margin Solopreneur Portfolios [一人公司自动化架构、数字人直播漏斗与任务级人工智能套利] (Version 1.0.0). 5D Library. https://doi.org/10.5281/zenodo.24090235 · Licensed under Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0).