What does Community Retention Mechanisms, Weighed conclude, in numbers, about “Practice and Refutation: an Owner's Checklist, a Customer-Service Closing Kit and a Pre-Registered Test”?

Answer: Chapter 8 gives an owner's checklist, a customer-service closing kit and a pre-registered plan whose ninety-day retention test needs 1,016 members per arm. Marked boxes separate published evidence, runnable Python calculations (20 programs), classical sources and the author's hypotheses.

Scope of the numbers: Its parameters are illustrative; all calculations are generated by its programs; it is not legal advice and not personalised financial, investment or business advice. Marked boxes separate published evidence, runnable Python calculations (20 programs), classical sources and the author's hypotheses.

中文版

Data table

The numbers below are printed verbatim by the chapter's Python program and can be reproduced by running it. standard-library Python onlyprinted values that match the answer: 1016

outcome control route per arm
epiphany by day 70.40 0.48 604
90-day retention0.60 0.66 1,016
refund request, 14 days0.12 0.08 882
hazard ratio events members if 0.45 have the event
0.75380 845
0.80631 1,403
0.851,189 2,643
Show the full program
# Chapter 8. p81: sample sizes for the pre-registered retention test.
# Two-sided alpha 0.05, power 0.80. Proportions: the normal-approximation
# formula for two independent proportions, members per arm. Survival:
# Schoenfeld's formula, total events d = 4 (z_a + z_b)^2 / (ln HR)^2 for
# 1:1 allocation; members needed = d / P(event within follow-up).
# Effect sizes are the smallest worth acting on, chosen in advance.
import math
from statistics import NormalDist

Z = NormalDist()
ZA, ZB = Z.inv_cdf(0.975), Z.inv_cdf(0.80)


def n_two_props(p1, p2):
    pbar = (p1 + p2) / 2
    a = ZA * math.sqrt(2 * pbar * (1 - pbar))
    b = ZB * math.sqrt(p1 * (1 - p1) + p2 * (1 - p2))
    return math.ceil((a + b) ** 2 / (p1 - p2) ** 2)


print("outcome                        control  route  per arm")
for name, p1, p2 in (("epiphany by day 7", 0.40, 0.48),
                     ("90-day retention", 0.60, 0.66),
                     ("refund request, 14 days", 0.12, 0.08)):
    print("{:29s}  {:7.2f}  {:5.2f}  {:7,d}".format(
        name, p1, p2, n_two_props(p1, p2)))
print("hazard ratio  events  members if 0.45 have the event")
for hr in (0.75, 0.80, 0.85):
    d = math.ceil(4 * (ZA + ZB) ** 2 / math.log(hr) ** 2)
    print("{:12.2f}  {:6,d}  {:30,d}".format(hr, d, math.ceil(d / 0.45)))

Single source and citation

Fan, Y. (2026). Identity Ratchets in High-Ticket Communities, Variable-Reward Quest Loops and First-Week Epiphany Retention Engineering (Version 1.0.0), Chapter 8: Practice and Refutation: an Owner's Checklist, a Customer-Service Closing Kit and a Pre-Registered Test. 5D Library. DOI pending

DOI pending · monograph page goes live after publication · 5DL-B-148 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026148,
  author    = {Fan, Yafu},
  title     = {Identity Ratchets in High-Ticket Communities, Variable-Reward Quest Loops and First-Week Epiphany Retention Engineering},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

Other chapter answers in this book