What does Click-to-Chat Closing and Attribution conclude, in numbers, about “The Length Boundary of Real-Person Voice Notes and One-Tap Payment in the Chat”?
Answer: A voice conveys thought and humanity better than text, but the longer a voice note runs, the fewer people hear it to the end. In the illustrative model the best length is 65 seconds and the net gain at 60 seconds is 0.999 of the best; across 135 parameter combinations the best length is 60 seconds or less in only 0.459 of cases, yet the median regret of a 60-second cap is just 0.0005. Sixty seconds is therefore not an optimum but a robust upper limit. Each step removed from a payment link gains the reciprocal of that step's pass rate; link validity should equal the real deadline, and a link valid for the full 72.00 hours yields payment before the deadline from 0.996 of intending buyers.
Scope of the numbers: Its conversion rates, latencies, reply rates, sample sizes and amounts are illustrative, and all calculations are generated by its programs. It is general business guidance, not legal or marketing-compliance advice, and promises no income.
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: 0.0005, 0.459, 135, 60
| settings 135, best length < | 60 s in 62 (0.459) |
| best length quartiles | 49.0 / 64.0 / 82.0 s |
| regret of a 60 s cap: median 0.0005, mean 0.0134, max 0.1105 | |
| worst case: scale 120 s, shape 1.2, theta 40 s, kappa 0.05, | |
| best length 154 s | |
| settings with regret above 0.05 | 11 |
Show the full program
# Chapter 6. p62: how robust is a 60-second cap on voice notes?
# Same model as p61. Sweep listener scale (45 to 120 s), Weibull shape
# (1.2 to 2.0), front-loading theta (15 to 40 s) and cut-off cost kappa
# (0.05 to 0.30): 135 settings. For each, find the best length on a
# one-second grid up to 180 s and the regret of capping notes at 60 s.
import itertools
import math
import statistics
DT = 0.5
def curve(scale, shape, theta, kappa, tmax=180):
terms, out = [], {}
surv = lambda t: math.exp(-((t / scale) ** shape))
steps_per_s = int(round(1 / DT))
for i in range(tmax * steps_per_s):
a, b = i * DT, (i + 1) * DT
fa = math.exp(-a / theta) / theta * surv(a)
fb = math.exp(-b / theta) / theta * surv(b)
terms.append((fa + fb) * DT / 2)
if (i + 1) % steps_per_s == 0:
T = (i + 1) // steps_per_s
out[T] = math.fsum(terms) - kappa * (1 - surv(T))
return out
bests, regrets, under = [], [], 0
grid = itertools.product((45, 60, 75, 90, 120), (1.2, 1.6, 2.0),
(15, 25, 40), (0.05, 0.15, 0.30))
worst = None
for scale, shape, theta, kappa in grid:
g = curve(scale, shape, theta, kappa)
tb = max((v, T) for T, v in g.items() if T >= 5)[1]
capped = max(v for T, v in g.items() if 5 <= T <= 60)
reg = 1 - capped / g[tb]
bests.append(tb)
regrets.append(reg)
under += tb <= 60
if worst is None or reg > worst[0]:
worst = (reg, scale, shape, theta, kappa, tb)
n = len(bests)
print("settings %d, best length <= 60 s in %d (%.3f)" % (n, under,
under / n))
q = statistics.quantiles(bests, n=4)
print("best length quartiles: %.1f / %.1f / %.1f s" % tuple(q))
print("regret of a 60 s cap: median %.4f, mean %.4f, max %.4f" % (
statistics.median(regrets), math.fsum(regrets) / n, max(regrets)))
print("worst case: scale %d s, shape %.1f, theta %d s, kappa %.2f," % (
worst[1], worst[2], worst[3], worst[4]))
print(" best length %d s" % worst[5])
print("settings with regret above 0.05: %d" % len(
[r for r in regrets if r > 0.05]))Single source and citation
Fan, Y. (2026). Click-to-Messenger Conversion Funnels, Chat Attribution and Voice-Note Closing in Southeast Asian Commerce (Version 1.0.0), Chapter 6: The Length Boundary of Real-Person Voice Notes and One-Tap Payment in the Chat. 5D Library. DOI pending
DOI pending · monograph page goes live after publication · 5DL-B-366 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026366,
author = {Fan, Yafu},
title = {Click-to-Messenger Conversion Funnels, Chat Attribution and Voice-Note Closing in Southeast Asian Commerce},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}Other chapter answers in this book
- Chapter 1 · Funnel Products and Halving Friction Loss: the Break-Even of Direct-to-Chat Ads
- Chapter 2 · The Decay Curve of First-Response Latency: Triage Before Staff
- Chapter 3 · Chat Attribution: Last Click, the Shapley Value and Severed Sources
- Chapter 4 · Increments and Sample Sizes: Why Most Increments Cannot Be Measured
- Chapter 5 · Three Labels and the Optimal Stop Point: the Mathematics of Consent-Based Follow-Up
- Chapter 7 · Classical Sources East and West: Silence in the Guiguzi, Watching Deeds in the Analects, and the Voice in Classical Rhetoric
- Chapter 8 · Actuarial Tools and Execution Checklist: the Weekly Chat Ledger, the Cadence Checker and Pre-Registered Tests