What does The Name-Number, Digit-Field and Brand-Naming Codex conclude, in numbers, about “Brand-Naming Rules: Trade Word, Trademark, Sound-Alikes, Domains, Dialects, Recall and Length”?

Answer: Chapter 7 gives eleven brand-naming rules: the numerology filter passes about half (0.5436 with the trade character fang); the real clash risk is sound, at least 0.1175 with 20,000 marks in a class and 0.8647 when names crowd onto 100 favoured syllables; lowest-cost rejections first cut hours per candidate from 1.019 to 0.812.

Scope of the numbers: It is not legal, investment or other professional 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. printed values that match the answer: 0.1175, 0.8647, 100, 20000

same characters space12,250,000
same sound with tones space1,690,000
same sound ignoring tones space160,000
M2,000 marks: P(clash) chars 0.0002 tones 0.0012 toneless 0.0124
M20,000 marks: P(clash) chars 0.0016 tones 0.0118 toneless 0.1175
M50,000 marks: P(clash) chars 0.0041 tones 0.0292 toneless 0.2684
names drawn from 100 favoured syllables: space 10,000; M20,000 -> P(clash) 0.8647
names drawn from50 favoured syllables: space 2,500; M = 20,000 -> P(clash) 0.9997
expected sound-alike pairs among 20,000 toneless names1,250
Show the full program
# Chapter 7. p71: Why sound collisions, not character collisions, sink a two-character brand.
# Inputs are round counts, stated as assumptions: about 400 Mandarin syllables without tone and 1,300
# with tone (common estimates of the syllable inventory), 3,500 characters in the common-use list.
# A new two-syllable name is compared with M marks already registered in the same class. If names were
# spread uniformly, P(at least one clash) = 1 - (1 - 1/space)^M; real names crowd onto a small set of
# auspicious syllables, so the uniform figure is a floor, not an estimate.
SYL0, SYL1, CHARS = 400, 1300, 3500

if __name__ == "__main__":
    spaces = [("same characters", CHARS ** 2), ("same sound with tones", SYL1 ** 2),
              ("same sound ignoring tones", SYL0 ** 2)]
    for lab, sp in spaces:
        print("   {:<26s} space {:>12,}".format(lab, sp))
    for m in (2000, 20000, 50000):
        row = ["{:.4f}".format(1 - (1 - 1 / sp) ** m) for _, sp in spaces]
        print("   M = {:>6,} marks: P(clash) chars {}  tones {}  toneless {}".format(m, *row))
    # crowding: if brand syllables are drawn from only the top k toneless syllables
    for k in (100, 50):
        sp = k ** 2
        print("   names drawn from {:3d} favoured syllables: space {:,}; M = 20,000 -> P(clash) {:.4f}"
              .format(k, sp, 1 - (1 - 1 / sp) ** 20000))
    m = 20000
    pairs = m * (m - 1) / 2 / SYL0 ** 2
    print("   expected sound-alike pairs among {:,} toneless names: {:,.0f}".format(m, pairs))

Single source and citation

Fan, Y. (2026). The Name-Number, Digit-Field and Brand-Naming Codex (Version 0.1.0), Chapter 7: Brand-Naming Rules: Trade Word, Trademark, Sound-Alikes, Domains, Dialects, Recall and Length. 5D Library. DOI pending

DOI pending · monograph page goes live after publication · 5DL-B-127 · v0.1.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026127,
  author    = {Fan, Yafu},
  title     = {The Name-Number, Digit-Field and Brand-Naming Codex},
  year      = {2026},
  publisher = {5D Library},
  edition   = {0.1.0},
  note      = {DOI pending}
}

Other chapter answers in this book