What does Creator Certification: Signals and Grades conclude, in numbers, about “The Badge Display Cascade: Thresholds, Informational Cascades, Audits and Revocation”?

Answer: Adoption depends on the distribution of thresholds across the whole population. Program p61 reproduces Granovetter's example, in which replacing one person in a hundred changes the outcome from everyone acting to one person acting; with a threshold spread of 0.05 the start must be 0.131, with 0.10 only 0.038. Program p62 shows that an informational cascade among brands locks in the wrong direction with probability 0.155 when private signals are correct with probability 0.7, and that an insufficiently accurate audit report pulls the probability of a correct cascade down from 0.845 to 0.750. Program p63 shows that a certificate with no audits has a steady-state precision of only 0.514, and that keeping it at 0.85 requires auditing at least 0.282 of holders each quarter.

Scope of the numbers: Its numbers of creators and brands, scoring errors, costs, premiums and audit parameters are illustrative, and all calculations are generated by its programs. It is general business guidance, not individual, legal, tax or investment advice.

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: 0.05, 0.10, 0.282, 0.514, 0.85

audit shareholdersprecisionrevoked per qtrauditsaudit cost
0.0078680.5140.000.0
0.0562220.65092.6311186.7
0.1055530.728128.3555333.2
0.2049770.812158.8995597.2
0.3047190.857172.41416849.4
0.5044780.903185.122391343.5
1.0042730.946195.942732564.0
smallest audit share for precision 0.75: 0.120
smallest audit share for precision 0.80: 0.181
smallest audit share for precision 0.85: 0.282
Show the full program
# Chapter 6. p63: audit and revocation keep the holder pool credible.
# Each quarter 400 new holders enter, 0.89 of them strong (the precision
# of p21 at pass rate 0.20, rounded). A strong holder slips below the
# standard with probability 0.04 a quarter; 0.05 of all holders leave on
# their own. A random share a of active holders is re-audited each
# quarter; an audit catches a slipped holder with sensitivity 0.85 and
# revokes the certificate. Steady state after 80 quarters. Audit cost
# 0.6 value units per audit. All parameters are illustrative.
import math

INFLOW, STRONG0, SLIP, EXIT, SENS, COST = 400, 0.89, 0.04, 0.05, 0.85, 0.6


def steady(a):
    good = bad = 0.0
    revoked = 0.0
    for _ in range(80):
        slipped = good * SLIP
        good = (good - slipped) * (1 - EXIT) + INFLOW * STRONG0
        caught = bad * a * SENS
        revoked = caught
        bad = (bad - caught) * (1 - EXIT) + slipped * (1 - EXIT) \
            + INFLOW * (1 - STRONG0)
    return good, bad, revoked


print("audit share  holders  precision  revoked per qtr  audits  audit cost")
best = None
for a in (0.0, 0.05, 0.10, 0.20, 0.30, 0.50, 1.00):
    g, b, rv = steady(a)
    total = g + b
    prec = g / total
    audits = a * total
    print("   %.2f       %5.0f     %.3f        %5.1f        %5.0f    %6.1f" % (
        a, total, prec, rv, audits, audits * COST))
for target in (0.75, 0.80, 0.85):
    a = next(k / 1000 for k in range(0, 1001)
             if steady(k / 1000)[0] / math.fsum(steady(k / 1000)[:2])
             >= target)
    print("smallest audit share for precision %.2f: %.3f" % (target, a))

Single source and citation

Fan, Y. (2026). Bootstrapping Industry Accreditation Standards, Spence Signalling Tiers and Inbound Creator Certification Magnetics (Version 1.0.0), Chapter 6: The Badge Display Cascade: Thresholds, Informational Cascades, Audits and Revocation. 5D Library. DOI pending

DOI pending · monograph page goes live after publication · 5DL-B-358 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026358,
  author    = {Fan, Yafu},
  title     = {Bootstrapping Industry Accreditation Standards, Spence Signalling Tiers and Inbound Creator Certification Magnetics},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

Other chapter answers in this book