What does Paid-Ad Scaling for Paywalls conclude, in numbers, about “Calibrating Server-Side Events: Deduplication, Net-Value Signals and the Report Inflation Factor”?
Answer: Chapter 5 covers consent, deduplication by event identifier and net-value reporting for server-side events: reporting only the icebreaker price yields an illustrative profit of -1,432 and reporting net value 552, and a randomised holdout measures an illustrative calibration factor of 0.615.
Scope of the numbers: Under illustrative parameters one icebreaker buyer on tarotmalaysia is worth 2.68 times the icebreaker price, and a first-order break-even stop rule captures only 0.222 of the attainable profit across forty synthetic creatives. Parameters are illustrative and all calculations are generated by the book's programs.
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: -1432, 552
| bait | start CPA 12.00 | true value per buyer | 6.94 | ||
| impulse start CPA 20.00 | true value per buyer | 15.90 | |||
| deep | start CPA 25.00 | true value per buyer | 34.26 | ||
| signal | bait | impulse | deep | buyers | profit RM |
| S1 gross icebreaker | 2,020 | 980 | 0 | 167.7 | -1,432 |
| S2 gross + upsell | 0 | 1,320 | 1,680 | 121.1 | 61 |
| S3 net calibrated | 0 | 0 | 3,000 | 103.7 | 552 |
Show the full program
# Chapter 5. p52: which value the server sends decides where the
# optimiser spends. Three illustrative audience pockets of one paywall
# (c_max buyers a day, k RM scale, icebreaker refund, upsell rate,
# deep-report refund, repeat value):
# bait cheap icebreakers, rare upsell
# impulse upsell often, but many deep reports are refunded
# deep dearer icebreakers, readers who keep the deep report
# Buyers per day from spend x: c_max (1 - exp(-x / k)). A stylised
# optimiser splits RM 3,000 a day in RM 20 chunks, each chunk to the
# pocket with the largest reported value gain. Signals:
# S1 icebreaker gross price only (RM 9.90 per purchase)
# S2 gross icebreaker plus gross upsell events (9.90 + u * 129)
# S3 net value after refunds, fees and upsell (p11 funnel)
# Profit uses the true net value. Parameters are illustrative.
import math
P1, P2, CB, CBF, RATE, FIX, TAX = 9.90, 129.0, 0.005, 15.0, 0.034, 0.5, 0.06
POCKETS = {"bait": (250.0, 3000.0, 0.28, 0.01, 0.04, 0.50),
"impulse": (300.0, 6000.0, 0.15, 0.14, 0.45, 1.00),
"deep": (400.0, 10000.0, 0.04, 0.18, 0.04, 7.00)}
NAMES = ("bait", "impulse", "deep")
BUDGET, CHUNK = 3000.0, 20.0
def order_net(price, refund, cost):
kept = price * (1 - refund - CB) * (1 - TAX)
return kept - (price * RATE + FIX) - CB * CBF - cost
def true_value(p):
cmax, k, rf, u, rf2, rep = POCKETS[p]
return math.fsum([order_net(P1, rf, 0.40),
u * order_net(P2, rf2, 1.50), rep])
def buyers(p, x):
cmax, k = POCKETS[p][0], POCKETS[p][1]
return cmax * (1.0 - math.exp(-x / k))
SIGNALS = {"S1 gross icebreaker": lambda p: P1,
"S2 gross + upsell": lambda p: P1 + POCKETS[p][3] * P2,
"S3 net calibrated": true_value}
if __name__ == "__main__":
for p in NAMES:
print("{:<7} start CPA {:>5.2f} true value per buyer {:>6.2f}"
.format(p, POCKETS[p][1] / POCKETS[p][0], true_value(p)))
print(" signal bait impulse deep buyers profit RM")
for name, sig in SIGNALS.items():
x = {p: 0.0 for p in NAMES}
for _ in range(int(BUDGET / CHUNK)):
gain = {p: sig(p) * (buyers(p, x[p] + CHUNK) - buyers(p, x[p]))
for p in NAMES}
best = max(NAMES, key=lambda q: gain[q])
x[best] += CHUNK
nb = math.fsum(buyers(p, x[p]) for p in NAMES)
prof = math.fsum(buyers(p, x[p]) * true_value(p) for p in NAMES)
print(" {:<19} {:>6,.0f} {:>8,.0f} {:>6,.0f} {:>7.1f} {:>10,.0f}"
.format(name, x["bait"], x["impulse"], x["deep"], nb,
prof - BUDGET))Single source and citation
Fan, Y. (2026). Paid-Ad Scaling for Diagnostic Paywalls: Meta and TikTok Stopping Rules, Server-Side CAPI Calibration and Hundred-Order Daily Unit Economics (Version 1.0.0), Chapter 5: Calibrating Server-Side Events: Deduplication, Net-Value Signals and the Report Inflation Factor. 5D Library. DOI pending
DOI pending · monograph page goes live after publication · 5DL-B-345 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026345,
author = {Fan, Yafu},
title = {Paid-Ad Scaling for Diagnostic Paywalls: Meta and TikTok Stopping Rules, Server-Side CAPI Calibration and Hundred-Order Daily Unit Economics},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}Other chapter answers in this book
- Chapter 1 · Funnel Unit Economics: Why First-Order Break-Even Is Not a Stop Line
- Chapter 2 · Stopping at Zero: the Rule of Three, Gamma-Poisson Posteriors and Spend Red Lines
- Chapter 3 · Signal, Noise and Sequential Tests: Small Budgets Can Only Call "Clearly Bad"
- Chapter 4 · Many Creatives on a Small Budget: Thompson Sampling, Even Splits and Screening Speed
- Chapter 6 · Control Laws for Scaling: Saturation, Carry-Over and Marginal Returns
- Chapter 7 · Classical Sources: the Sunzi on Waging War and Hopkins's Scientific Advertising
- Chapter 8 · Closing the Loop: a Three-Site Ad Ledger, the Distilled Tool and a Pre-Registered Refutation