What does First-Minute Onboarding and Stickiness conclude, in numbers, about “Feedback Within a Hundred Milliseconds: Latency Thresholds Beat Lavish Effects”?

Answer: Latency is a distribution, and the battle is fought in its tail. Two versions whose medians differ by only 0.35 seconds can differ by more than a factor of ten in the share of responses above one second. A skeleton screen shortens no latency, yet in the illustrative model it reduces the share of visitors who leave by 0.25; optimistic updates suit only operations that rarely fail and can be fully restored. Serial requests push the P95 over one second quickly, and with these parameters no more than 2 serial requests fit before the first result. Effects are welcome, but they come after latency, and they can always be switched off.

Scope of the numbers: Enlarging a tap target from 24 to 44 pixels lowers the illustrative miss rate from 0.0865 to 0.0017. Parameters are illustrative and correspond to no real user or sale; all calculations are generated by the book's programs.

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: 0.25, 95

limits in seconds0.1, 1.0, 10.0; Doherty line 0.4
profile P50 P95 sigma <0.10.1-1 1-10 >10 <0.4
A local echo0.06 0.15 0.557 0.820 0.180 0.000 0.000 1.000
B fast API0.25 0.80 0.707 0.098 0.878 0.025 0.000 0.747
C slow API0.60 2.40 0.843 0.017 0.711 0.272 0.000 0.315
D cold start1.20 9.00 1.225 0.021 0.420 0.517 0.042 0.185
P95 budget check: share above 1.0 s must stay below 0.05
A local echo above 1 s 0.000 ok
B fast API above 1 s 0.025 ok
C slow API above 1 s 0.272 RED
D cold start above 1 s 0.559 RED
Show the full program
# Chapter 2. p21: where a lognormal latency falls among the 0.1 s,
# 1 s and 10 s limits. A latency profile is set by its median P50 and
# its 95th percentile P95; sigma = ln(P95 / P50) / 1.6449. Shares come
# from the normal CDF (math.erf). The four profiles are illustrative.
import math

Z95 = 1.6449


def cdf(x, mu, sigma):
    return 0.5 * (1.0 + math.erf((math.log(x) - mu) / (sigma * 2 ** 0.5)))


def bands(p50, p95):
    mu, sigma = math.log(p50), math.log(p95 / p50) / Z95
    c1, c2, c3 = (cdf(x, mu, sigma) for x in (0.1, 1.0, 10.0))
    c4 = cdf(0.4, mu, sigma)
    return sigma, c1, c2 - c1, c3 - c2, 1.0 - c3, c4


if __name__ == "__main__":
    profiles = [("A local echo", 0.06, 0.15), ("B fast API", 0.25, 0.80),
                ("C slow API", 0.60, 2.40), ("D cold start", 1.20, 9.00)]
    print("limits in seconds: 0.1, 1.0, 10.0; Doherty line 0.4")
    print("profile        P50   P95  sigma  <0.1  0.1-1  1-10   >10"
          "  <0.4")
    for name, p50, p95 in profiles:
        s, a, b, c, d, e = bands(p50, p95)
        print("{:<13} {:>4.2f} {:>5.2f}  {:.3f} {:.3f}  {:.3f} {:.3f}"
              " {:.3f} {:.3f}".format(name, p50, p95, s, a, b, c, d, e))
    print("P95 budget check: share above 1.0 s must stay below 0.05")
    for name, p50, p95 in profiles:
        s, a, b, c, d, e = bands(p50, p95)
        print("  {:<13} above 1 s {:.3f}  {}".format(
            name, c + d, "ok" if c + d < 0.05 else "RED"))

Single source and citation

Fan, Y. (2026). Zero-Friction First-Minute Onboarding, Micro-Interaction Habit Loops and High-Stickiness Interface Engineering (Version 1.0.0), Chapter 2: Feedback Within a Hundred Milliseconds: Latency Thresholds Beat Lavish Effects. 5D Library. DOI pending

DOI pending · monograph page goes live after publication · 5DL-B-379 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026379,
  author    = {Fan, Yafu},
  title     = {Zero-Friction First-Minute Onboarding, Micro-Interaction Habit Loops and High-Stickiness Interface Engineering},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

Other chapter answers in this book