中文

Daily campaign report — 2026-08-20, against 2026-08-19

The four ad accounts spent $608.21 on 2026-08-20 against $509.12 the day before, and for the first time the money went to two products: $575.05 on Vloom and $33.16 on Sinai, which launched into B1 and HR1 during the day. Vloom bought 667 registrations against 639 and sealed $319.31 inside the day against $597.16, a 46.5% fall on 12.9% more spend. Sinai bought 117 and sealed $68.15, a 2.06 return on its first day. A one-day United States test in KBM1 took a further $25.91 and is excluded from every Sinai figure on this page; it is reported on its own below. Definitions, the report clock and the three revenue measures are in the appendix.

1Overview

(i) Both apps, both days. Spend is additive across the two products; revenue is two ledgers, Vloom's the Lum feed and Sinai's the IAP feed.

2026-08-192026-08-20change
spend$509.12$608.21+19.5%
impressions57,12957,152+0.0%
CPM$8.91$10.64+19.4%
registrations (ours)639784+22.7%
cost per registration$0.797$0.776−2.6%
booked$1,866.87 (3.67)$2,147.48 (3.53)+15.0%
Meta booked$1,544.12 (3.03)$1,840.35 (3.03)
within-day$597.16 (1.17)$387.46 (0.64)−35.1%
cohort @ 24h$863.62 (1.70)$488.71 (0.80)−43.4%
booked payers143163+20
payer rate22.4%20.8%−1.6 pts
ARPU$2.922$2.739−6.2%
ARPPU$13.06$13.17+0.9%

Cohort value — ARPU and ARPPU at d0, d3 and d7

Keyed on the INSTALL day and measured from each user's own install stamp: dN is their first (N+1) x 24 h. ARPU is Adjust's own series, the one the partner's dashboard reports; ARPPU is our payment ledger, because Adjust exposes no cumulative distinct-payer count. A horizon the window has not lived through yet is left empty rather than written as zero.

appARPU (d0)ARPU (d3)ARPU (d7)ARPPU (d0)ARPPU (d3)ARPPU (d7)installs
Vloom$0.4935$0.7613$0.8512$7.34$8.93$9.44610
Sinai AI$0.9535$2.1031$2.9616$7.33$13.61$15.84123

Vloom is $575.05 of that and Sinai $33.16: 12.9% more on the running product, plus a new one starting.

Sinai's first day is 5.5% of the portfolio's spend and 17.6% of its sealed return, bought at $0.283 a registration against Vloom's $0.862.

The one-day United States test is why Sinai reads 2.06 and not 1.26: pooling it reports a geography decision as the app's result.

The hour-neutral measure widens the gap: Sinai 3.43 against Vloom 0.65 is 5.3×, where the sealed measure shows 3.7.

Vloom's sealed return halved — the series' largest one-day fall, and cohort@24h puts it at 56.6%. Inside B1: 1.28 to 0.70 on 6.5% more spend.

The two US-only accounts are the expensive quarter of the buy. $128.94 of Vloom's $575.05, at CPMs of $161.74 and $29.16 against B1's $8.35.

2The two apps, one tab each

Each tab below is that app's whole page: its accounts, then its hours, its countries, its aggregate, its campaigns and its assets, in that order and on its own ledger. Every panel is written into the page and the tab only hides it, so with scripting off both apps are present and in order.

Showing

Vloom — every section below this line is Vloom's: the day, the hours, the countries, the campaigns and the assets, counted on Vloom's own ledger and its own accounts. Use the buttons above to switch.

The day, and every account

(i) Vloom, both days.

2026-08-192026-08-20change
spend$509.12$575.05+12.9%
impressions57,12955,440−3.0%
CPM$8.91$10.37+16.4%
registrations (ours)639667+4.4%
cost per registration$0.797$0.862+8.2%
booked$1,866.87 (3.67)$2,079.32 (3.62)+11.4%
Meta booked$1,544.12 (3.03)$1,762.52 (3.06)
within-day$597.16 (1.17)$319.31 (0.56)−46.5%
cohort @ 24h$863.62 (1.70)$375.07 (0.65)−56.6%
booked payers143151+8
payer rate (d0)8.6%6.1%−2.5 pts
ARPU (d0)$1.3515$0.5623−58.4%
ARPPU (d0)$15.70$9.15−41.7%

(ii) Vloom, both days, every account.

2026-08-20 (the day) — delivery

accountspendshareimpressionsCPMregistrationscost/reg
Portfolio$575.05100%55,440$10.37667$0.862
B1$425.8174%50,967$8.35560$0.760
Nomad Node$20.304%2,270$8.9441$0.495
KBM1$78.9314%488$161.7419$4.154
HR1$50.019%1,715$29.1647$1.064

2026-08-20 (the day) — the three measures

accountbookedbROASwithin-daywROAScohort@24hcROASpayers (d0)payer% (d0)ARPU (d0)ARPPU (d0)
Portfolio$2,079.323.62$319.310.56$375.070.65416.1%$0.5623$9.15
B1$1,820.744.28$297.590.70$353.350.83386.8%$0.6310$9.30
Nomad Node$196.889.70$0.000.00$0.000.0000.0%$0.0000—
KBM1$55.950.71$15.960.20$15.960.20210.5%$0.8400$7.98
HR1$5.760.12$5.760.12$5.760.1212.1%$0.1225$5.76

Each account, 2026-08-19 against 2026-08-20, one table each.

B1

2026-08-192026-08-20change
spend$399.95$425.81+6.5%
impressions54,33050,967−6.2%
CPM$7.36$8.35+13.5%
registrations (ours)596560−6.0%
cost per registration$0.671$0.760+13.3%
booked$1,475.97 (3.69)$1,820.74 (4.28)+23.4%
Meta booked$1,146.72 (2.87)$1,410.70 (3.31)
within-day$511.92 (1.28)$297.59 (0.70)−41.9%
cohort @ 24h$778.37 (1.95)$353.35 (0.83)−54.6%
booked payers121131+10
payer rate (d0)8.4%6.8%−1.6 pts
ARPU (d0)$1.3060$0.6310−51.7%
ARPPU (d0)$15.57$9.30−40.3%

Nomad Node

2026-08-192026-08-20change
spend$20.31$20.30−0.0%
impressions2,1752,270+4.4%
CPM$9.34$8.94−4.2%
registrations (ours)2841+46.4%
cost per registration$0.725$0.495−31.7%
booked$292.97 (14.42)$196.88 (9.70)−32.8%
Meta booked$294.47 (14.50)$295.80 (14.57)
within-day$27.29 (1.34)$0.00 (0.00)−100.0%
cohort @ 24h$27.29 (1.34)$0.00 (0.00)−100.0%
booked payers1816−2
payer rate (d0)7.1%0.0%−7.1 pts
ARPU (d0)$0.9748$0.0000−100.0%
ARPPU (d0)$13.65$0.00−100.0%

KBM1

2026-08-192026-08-20change
spend$63.03$78.93+25.2%
impressions358488+36.3%
CPM$176.06$161.74−8.1%
registrations (ours)1119+72.7%
cost per registration$5.730$4.154−27.5%
booked$92.95 (1.47)$55.95 (0.71)−39.8%
Meta booked$97.94 (1.55)$46.93 (0.59)
within-day$52.96 (0.84)$15.96 (0.20)−69.9%
cohort @ 24h$52.96 (0.84)$15.96 (0.20)−69.9%
booked payers33+0
payer rate (d0)18.2%10.5%−7.7 pts
ARPU (d0)$4.8145$0.8400−82.6%
ARPPU (d0)$26.48$7.98−69.9%

HR1

2026-08-192026-08-20change
spend$25.83$50.01+93.6%
impressions2661,715+544.7%
CPM$97.11$29.16−70.0%
registrations (ours)447+1075.0%
cost per registration$6.458$1.064−83.5%
booked$4.99 (0.19)$5.76 (0.12)+15.4%
Meta booked$4.99 (0.19)$9.09 (0.18)
within-day$4.99 (0.19)$5.76 (0.12)+15.4%
cohort @ 24h$4.99 (0.19)$5.76 (0.12)+15.4%
booked payers11+0
payer rate (d0)25.0%2.1%−22.9 pts
ARPU (d0)$1.2475$0.1225−90.2%
ARPPU (d0)$4.99$5.76+15.4%

Hourly, 2026-08-20

(i) The portfolio, hour by hour, 2026-08-20. Every account placeable on the report clock, summed; delivery on the delivery hour, our columns on the hour the appendix names.

hrspendimpressionsCPMregistrationscost/regbookedbROASwithin-daywROAScohort@24hcROASbooked payers
00$25.632,292$11.1822$1.165$103.534.04$40.721.59$40.721.5913
01$23.321,923$12.1323$1.014$73.103.13$16.730.72$16.730.7210
02$33.073,227$10.2533$1.002$30.420.92$3.400.10$3.400.105
03$37.123,937$9.4347$0.790$205.625.54$0.000.00$15.180.416
04$24.893,214$7.7434$0.732$93.083.74$19.960.80$19.960.806
05$27.632,919$9.4725$1.105$14.940.54$5.010.18$5.010.184
06$24.743,006$8.2336$0.687$69.052.79$0.000.00$0.000.007
07$29.982,621$11.4429$1.034$86.582.89$4.990.17$4.990.178
08$20.712,650$7.8219$1.090$100.194.84$0.000.00$0.000.005
09$22.832,555$8.9429$0.787$133.895.86$0.000.00$0.000.0011
10$26.952,832$9.5234$0.793$36.551.36$43.161.60$43.161.607
11$25.882,799$9.2528$0.924$59.222.29$11.030.43$11.030.436
12$27.232,652$10.2733$0.825$89.973.30$14.710.54$14.710.548
13$19.282,397$8.0418$1.071$37.661.95$5.760.30$5.760.307
14$23.671,494$15.8428$0.845$125.295.29$9.980.42$9.980.4210
15$28.882,542$11.3636$0.802$83.962.91$10.970.38$10.970.3812
16$27.492,451$11.2227$1.018$135.784.94$46.511.69$46.511.6912
17$18.571,611$11.5327$0.688$79.254.27$18.621.00$18.621.0013
18$17.021,323$12.8622$0.774$123.617.26$0.000.00$0.000.009
19$16.661,357$12.2829$0.574$47.542.85$0.000.00$0.000.005
20$19.861,545$12.8520$0.993$96.204.84$62.073.13$62.073.1311
21$18.141,449$12.5223$0.789$137.337.57$0.000.00$14.230.788
22$19.001,320$14.3922$0.864$27.551.45$0.000.00$15.980.845
23$16.501,324$12.4623$0.717$89.005.39$5.710.35$16.080.978

(ii) The same 24 hours per account, 2026-08-20.

B1 — $425.81 on this day

hrspendimpressionsCPMregistrationscost/regbookedbROASwithin-daywROAScohort@24hcROASbooked payers
00$15.642,204$7.1020$0.782$103.536.62$40.722.60$40.722.6013
01$12.571,831$6.8719$0.662$65.125.18$5.760.46$5.760.469
02$16.672,212$7.5416$1.042$24.441.47$3.400.20$3.400.203
03$21.823,171$6.8839$0.559$205.629.42$0.000.00$15.180.706
04$18.342,834$6.4726$0.705$88.094.80$14.970.82$14.970.825
05$20.982,547$8.2420$1.049$14.940.71$5.010.24$5.010.244
06$20.362,767$7.3628$0.727$69.053.39$0.000.00$0.000.007
07$18.062,439$7.4026$0.695$86.584.79$4.990.28$4.990.288
08$16.912,524$6.7018$0.939$100.195.93$0.000.00$0.000.005
09$17.302,401$7.2123$0.752$54.173.13$0.000.00$0.000.007
10$24.412,755$8.8632$0.763$33.561.37$43.161.77$43.161.776
11$21.232,682$7.9227$0.786$43.112.03$11.030.52$11.030.525
12$25.002,586$9.6729$0.862$89.973.60$14.710.59$14.710.598
13$17.062,336$7.3017$1.004$37.662.21$5.760.34$5.760.347
14$19.881,418$14.0227$0.736$103.485.21$9.980.50$9.980.508
15$21.412,446$8.7533$0.649$54.872.56$10.970.51$10.970.518
16$20.582,329$8.8423$0.895$92.504.49$40.751.98$40.751.9810
17$13.821,499$9.2223$0.601$70.505.10$18.621.35$18.621.3511
18$13.901,238$11.2318$0.772$117.668.46$0.000.00$0.000.007
19$11.961,303$9.1824$0.498$44.583.73$0.000.00$0.000.004
20$16.411,487$11.0418$0.912$70.194.28$62.073.78$62.073.7810
21$15.551,396$11.1419$0.818$137.338.83$0.000.00$14.230.918
22$12.081,271$9.5012$1.007$24.592.04$0.000.00$15.981.324
23$13.871,291$10.7423$0.603$89.006.42$5.710.41$16.081.168

Nomad Node — $20.30 on this day

hrspendimpressionsCPMregistrationscost/regbookedbROASwithin-daywROAScohort@24hcROASbooked payers
00$0.000—0—$0.00—$0.00—$0.00—0
01$0.0112$0.830—$0.000.00$0.000.00$0.000.000
02$8.01808$9.9112$0.667$2.990.37$0.000.00$0.000.001
03$6.02574$10.496$1.003$0.000.00$0.000.00$0.000.000
04$2.34303$7.726$0.390$0.000.00$0.000.00$0.000.000
05$1.72252$6.833$0.573$0.000.00$0.000.00$0.000.000
06$1.08153$7.067$0.154$0.000.00$0.000.00$0.000.000
07$0.89107$8.321$0.890$0.000.00$0.000.00$0.000.000
08$0.1429$4.830—$0.000.00$0.000.00$0.000.000
09$0.0523$2.171$0.050$79.721594.41$0.000.00$0.000.004
10$0.049$4.441$0.040$2.9974.83$0.000.00$0.000.001
11$0.000—0—$16.10—$0.00—$0.00—1
12$0.000—1$0.000$0.00—$0.00—$0.00—0
13$0.000—0—$0.00—$0.00—$0.00—0
14$0.000—0—$21.81—$0.00—$0.00—2
15$0.000—0—$29.09—$0.00—$0.00—4
16$0.000—1$0.000$3.29—$0.00—$0.00—1
17$0.000—0—$2.99—$0.00—$0.00—1
18$0.000—1$0.000$5.95—$0.00—$0.00—2
19$0.000—0—$2.96—$0.00—$0.00—1
20$0.000—0—$26.02—$0.00—$0.00—1
21$0.000—1$0.000$0.00—$0.00—$0.00—0
22$0.000—0—$2.96—$0.00—$0.00—1
23$0.000—0—$0.00—$0.00—$0.00—0

KBM1 — $78.93 on this day

hrspendimpressionsCPMregistrationscost/regbookedbROASwithin-daywROAScohort@24hcROASbooked payers
00$6.7460$112.332$3.370$0.000.00$0.000.00$0.000.000
01$8.6459$146.443$2.880$7.980.92$10.971.27$10.971.271
02$5.0651$99.222$2.530$2.990.59$0.000.00$0.000.001
03$5.8841$143.410—$0.000.00$0.000.00$0.000.000
04$3.1717$186.471$3.170$4.991.57$4.991.57$4.991.571
05$1.7815$118.670—$0.000.00$0.000.00$0.000.000
06$1.2314$87.860—$0.000.00$0.000.00$0.000.000
07$9.8033$296.972$4.900$0.000.00$0.000.00$0.000.000
08$1.9318$107.220—$0.000.00$0.000.00$0.000.000
09$3.3522$152.271$3.350$0.000.00$0.000.00$0.000.000
10$0.918$113.750—$0.000.00$0.000.00$0.000.000
11$2.0312$169.170—$0.000.00$0.000.00$0.000.000
12$0.727$102.860—$0.000.00$0.000.00$0.000.000
13$1.2812$106.671$1.280$0.000.00$0.000.00$0.000.000
14$1.5312$127.500—$0.000.00$0.000.00$0.000.000
15$5.6715$378.000—$0.000.00$0.000.00$0.000.000
16$4.0214$287.140—$39.999.95$0.000.00$0.000.001
17$2.6914$192.140—$0.000.00$0.000.00$0.000.000
18$1.367$194.290—$0.000.00$0.000.00$0.000.000
19$3.2712$272.500—$0.000.00$0.000.00$0.000.000
20$1.7111$155.451$1.710$0.000.00$0.000.00$0.000.000
21$0.9510$95.001$0.950$0.000.00$0.000.00$0.000.000
22$3.9914$285.005$0.798$0.000.00$0.000.00$0.000.000
23$1.2210$122.000—$0.000.00$0.000.00$0.000.000

HR1 — $50.01 on this day

hrspendimpressionsCPMregistrationscost/regbookedbROASwithin-daywROAScohort@24hcROASbooked payers
00$3.2528$116.070—$0.000.00$0.000.00$0.000.000
01$2.1021$100.001$2.100$0.000.00$0.000.00$0.000.000
02$3.33156$21.353$1.110$0.000.00$0.000.00$0.000.000
03$3.40151$22.522$1.700$0.000.00$0.000.00$0.000.000
04$1.0460$17.331$1.040$0.000.00$0.000.00$0.000.000
05$3.15105$30.002$1.575$0.000.00$0.000.00$0.000.000
06$2.0772$28.751$2.070$0.000.00$0.000.00$0.000.000
07$1.2342$29.290—$0.000.00$0.000.00$0.000.000
08$1.7379$21.901$1.730$0.000.00$0.000.00$0.000.000
09$2.13109$19.544$0.532$0.000.00$0.000.00$0.000.000
10$1.5960$26.501$1.590$0.000.00$0.000.00$0.000.000
11$2.62105$24.951$2.620$0.000.00$0.000.00$0.000.000
12$1.5159$25.593$0.503$0.000.00$0.000.00$0.000.000
13$0.9449$19.180—$0.000.00$0.000.00$0.000.000
14$2.2664$35.311$2.260$0.000.00$0.000.00$0.000.000
15$1.8081$22.223$0.600$0.000.00$0.000.00$0.000.000
16$2.89108$26.763$0.963$0.000.00$5.761.99$5.761.990
17$2.0698$21.024$0.515$5.762.80$0.000.00$0.000.001
18$1.7678$22.563$0.587$0.000.00$0.000.00$0.000.000
19$1.4342$34.055$0.286$0.000.00$0.000.00$0.000.000
20$1.7447$37.021$1.740$0.000.00$0.000.00$0.000.000
21$1.6443$38.142$0.820$0.000.00$0.000.00$0.000.000
22$2.9335$83.715$0.586$0.000.00$0.000.00$0.000.000
23$1.4123$61.300—$0.000.00$0.000.00$0.000.000
2026-08-21T18:23:58.066445 image/svg+xml Matplotlib v3.11.1, https://matplotlib.org/ 0 3 6 9 12 15 18 21 hour (UTC) 0 5 10 15 20 25 USD Spend by hour 0 3 6 9 12 15 18 21 hour (UTC) 1250 1500 1750 2000 2250 2500 2750 3000 3250 impressions Impressions and installs by hour impressions installs 10.0 12.5 15.0 17.5 20.0 22.5 25.0 27.5 installs (Meta Leads) The day's motion — whole account, 2026-08-20 (UTC)
The day's motion: spend by hour, and impressions against installs
2026-08-21T18:23:58.205105 image/svg+xml Matplotlib v3.11.1, https://matplotlib.org/ 0 4 8 12 16 20 hour (UTC) 5 6 7 8 9 10 CTR % CTR 0 4 8 12 16 20 hour (UTC) 7 8 9 10 11 12 13 14 USD per 1,000 impressions CPM 0 4 8 12 16 20 hour (UTC) 6 8 10 12 14 16 18 20 22 installs / clicks % Click → install 0 4 8 12 16 20 hour (UTC) 0.8 1.0 1.2 1.4 1.6 USD Cost per install Delivery by hour — whole account, 2026-08-20 (UTC)
CTR, CPM, click-to-install and CPI by hour, whole account

Against the prior day, hour matched to hour

Hours both days delivered in, on the UTC clock: 00, 01, 02, 03, 04, 05, 06, 07, 08, 09, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23. Only the shared hours are compared.

Counting which way each shared hour points, which needs no weighting and asks only for a direction:

metrichours 2026-08-20 ran higherhours it ran lowersign test
CPM18 of 246 of 24p = 0.023
CTR12 of 2412 of 24p = 1.000
click→install13 of 2411 of 24p = 0.839
CPI14 of 2410 of 24p = 0.541

A sign test throws magnitude away and buys weight-free direction with it.

The full hour-by-hour pairing, 2026-08-19 → 2026-08-20

Every shared hour, 2026-08-19 → 2026-08-20. All columns are Meta's instrument on both sides of each ratio. Revenue is deliberately absent: per-hour payer counts run 0-3, so an hourly revenue column would be one person's basket read as a rate.

hrimpressionsCPMCTRclick→installCPI
001,776 → 2,2047.59 → 7.104.73% → 5.58%9.5% → 15.4%1.685 → 0.823
011,518 → 1,8317.54 → 6.875.40% → 6.28%13.4% → 14.8%1.041 → 0.739
021,975 → 2,2127.75 → 7.545.97% → 5.11%17.8% → 12.4%0.729 → 1.191
032,465 → 3,1717.18 → 6.885.76% → 6.34%14.1% → 11.9%0.885 → 0.909
043,101 → 2,8345.56 → 6.475.16% → 5.15%11.9% → 15.1%0.908 → 0.834
052,879 → 2,5476.56 → 8.245.90% → 6.12%15.3% → 10.9%0.726 → 1.234
062,082 → 2,7675.89 → 7.364.71% → 5.53%14.3% → 16.3%0.876 → 0.814
071,164 → 2,4399.89 → 7.408.16% → 4.67%9.5% → 19.3%1.279 → 0.821
081,552 → 2,5246.45 → 6.706.06% → 5.35%19.1% → 9.6%0.556 → 1.301
092,183 → 2,4016.85 → 7.214.76% → 4.71%10.6% → 18.6%1.359 → 0.824
102,153 → 2,7556.60 → 8.865.16% → 6.28%13.5% → 16.2%0.948 → 0.872
112,684 → 2,6827.26 → 7.925.33% → 6.79%14.7% → 11.0%0.928 → 1.062
122,369 → 2,5866.12 → 9.676.08% → 7.00%4.9% → 13.3%2.071 → 1.042
132,204 → 2,3367.73 → 7.307.26% → 6.81%10.6% → 6.3%1.002 → 1.706
141,410 → 1,41813.94 → 14.0210.14% → 10.44%13.3% → 9.5%1.034 → 1.420
152,252 → 2,4467.48 → 8.757.42% → 8.26%9.6% → 8.9%1.053 → 1.189
162,528 → 2,3297.67 → 8.846.84% → 9.02%13.3% → 10.0%0.843 → 0.980
173,349 → 1,4998.23 → 9.227.35% → 7.27%12.2% → 11.0%0.919 → 1.152
183,056 → 1,2386.58 → 11.236.32% → 5.82%13.0% → 16.7%0.804 → 1.158
192,587 → 1,3038.21 → 9.186.30% → 6.14%12.3% → 21.2%1.062 → 0.704
202,458 → 1,4878.28 → 11.047.08% → 5.85%10.9% → 12.6%1.071 → 1.492
212,439 → 1,3967.79 → 11.146.19% → 5.44%12.6% → 15.8%0.999 → 1.296
222,017 → 1,2716.39 → 9.505.70% → 5.51%19.1% → 12.9%0.586 → 1.342
232,129 → 1,2916.99 → 10.745.40% → 5.89%15.7% → 22.4%0.827 → 0.816
2026-08-21T18:23:58.356915 image/svg+xml Matplotlib v3.11.1, https://matplotlib.org/ 0 5 10 15 20 hour (UTC) 5 6 7 8 9 10 CTR % CTR 2026-08-19 2026-08-20 0 5 10 15 20 hour (UTC) 6 8 10 12 14 USD per 1,000 impressions CPM 0 5 10 15 20 hour (UTC) 5.0 7.5 10.0 12.5 15.0 17.5 20.0 22.5 installs / clicks % Click → install 0 5 10 15 20 hour (UTC) 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 USD Cost per install Delivery by hour — 2026-08-20 against 2026-08-19, shared hours only (UTC)
The same four ratios with the prior day laid over the day, shared hours only

CPM rose in 18 of 24 hours, p = 0.023. 12.8% more spend bought 3.0% fewer impressions at 16.3% more per thousand; nothing else separated.

Country, 2026-08-20

2026-08-23T12:43:23.113753 image/svg+xml Matplotlib v3.11.1, https://matplotlib.org/ US IN ID GB MX TR PH MY AR CO CL CA 0 20 40 60 80 100 120 140 160 USD delivery raked onto the report clock against two measured margins; revenue is the report day Spend against within-day revenue by country — 2026-08-20 spend within-day revenue
Spend against within-day revenue, by country
2026-08-23T12:43:23.163269 image/svg+xml Matplotlib v3.11.1, https://matplotlib.org/ −0.04 −0.02 0.00 0.02 0.04 USD per install Cost per install by country — 2026-08-20 (report clock)
Cost per install by country

(i) The portfolio by country, 2026-08-20. On the report clock. Meta serves no hour x country grid, so each country cut is raked from its own Meta day onto UTC; our own columns were already there.

ccspendshareimpressionsCPMregistrations (ours)cost/regwithin-daywROASpayers
US$165.2128.7%3,094$53.4070$2.360$133.780.8131
IN$101.0717.6%12,091$8.36162$0.624$46.590.4619
ID$19.423.4%3,261$5.9530$0.647$22.671.179
GB$18.333.2%1,039$17.6515$1.222$6.040.336
MX$15.892.8%3,360$4.7330$0.530$9.110.579
TR$15.272.7%950$16.0717$0.899$0.000.004
PH$13.402.3%5,922$2.2626$0.516$0.000.002
MY$12.172.1%2,247$5.4224$0.507$5.380.443
AR$10.231.8%973$10.5117$0.602$0.000.002
CO$9.581.7%1,022$9.3716$0.598$0.000.001
CL$9.291.6%728$12.7513$0.714$0.000.002
CA$8.921.6%452$19.746$1.487$4.990.562

189 further countries are not listed, $176.38 between them (30.7% of this block).

Reconciliation, what is measured and what is fitted. Every per-country cell is an allocation across hours; every total is measured. B1 (UTC-7) $126.38 from Meta 2026-08-19 and $299.51 from Meta 2026-08-20, a measured UTC-day total of $425.89; Nomad Node (UTC+0) needed none, its Meta day IS the UTC day at $20.30; KBM1 (UTC-7) $32.50 from Meta 2026-08-19 and $46.43 from Meta 2026-08-20, a measured UTC-day total of $78.93; HR1 (UTC+0) needed none, its Meta day IS the UTC day at $50.04. The 2 UTC+0 accounts above are the control: raking one returns its country export unchanged, and that identity held here. Meta days this run did not pull were taken from the country archive at ad_ops/_data/country_archive/ — KBM1 Meta 2026-08-19 — because a UTC-7 day reaches back into the Meta day before it and the run window did not.

(ii) The same cut per account, 2026-08-20.

B1

ccspendshareimpressionsCPMregistrations (ours)cost/regwithin-daywROASpayers
IN$85.7320.1%10,810$7.93140$0.612$46.590.5419
US$63.9615.0%2,378$26.9043$1.487$117.821.8426
GB$17.844.2%1,024$17.4315$1.190$6.040.346
ID$16.773.9%2,809$5.9722$0.762$22.671.357
MX$14.773.5%3,229$4.5726$0.568$9.110.628
PH$12.312.9%5,736$2.1526$0.474$0.000.002
TR$12.172.9%892$13.6413$0.936$0.000.003
MY$10.572.5%2,157$4.9024$0.440$5.380.513
AR$9.172.2%889$10.3115$0.611$0.000.001
CO$8.842.1%969$9.1212$0.736$0.000.001
CA$8.532.0%442$19.306$1.422$4.990.582
CL$8.512.0%683$12.4512$0.709$0.000.002

187 further countries are not listed, $156.72 between them (36.8% of this block).

Nomad Node

ccspendshareimpressionsCPMregistrations (ours)cost/regwithin-daywROASpayers
IN$3.2616.1%520$6.275$0.652$0.000.000
ID$1.557.6%371$4.188$0.194$0.000.002
US$1.416.9%13$108.461$1.410$0.000.002
TH$1.085.3%143$7.551$1.080$0.000.002
MY$1.004.9%69$14.490—$0.000.000
BR$0.793.9%84$9.401$0.790$0.000.001
MX$0.643.2%118$5.423$0.213$0.000.001
KR$0.613.0%22$27.733$0.203$0.000.000
PH$0.552.7%155$3.550—$0.000.000
PT$0.552.7%5$110.000—$0.000.000
UA$0.502.5%16$31.250—$0.000.000
TW$0.492.4%29$16.900—$0.000.001

85 further countries are not listed, $7.87 between them (38.8% of this block).

KBM1

ccspendshareimpressionsCPMregistrations (ours)cost/regwithin-daywROASpayers
US$78.93100.0%488$161.7419$4.154$15.960.203

HR1

ccspendshareimpressionsCPMregistrations (ours)cost/regwithin-daywROASpayers
US$20.9141.8%215$97.267$2.987$0.000.000
IN$12.0824.1%761$15.8817$0.711$0.000.000
BR$2.885.8%86$33.467$0.411$0.000.000
TR$2.675.3%42$63.542$1.334$0.000.000
ID$1.102.2%81$13.570—$0.000.000
AU$0.701.4%15$46.630—$0.000.000
TH$0.671.3%46$14.551$0.670$0.000.000
MY$0.601.2%21$28.550—$0.000.000
AR$0.581.2%31$18.691$0.580$0.000.000
PH$0.541.1%31$17.400—$0.000.000
MX$0.481.0%13$36.901$0.480$0.000.000
PR$0.390.8%3$129.961$0.390$0.000.000

73 further countries are not listed, $6.45 between them (12.9% of this block).

The prior day, 2026-08-19, the same two tables.

(i) The portfolio by country, 2026-08-19. On the report clock. Meta serves no hour x country grid, so each country cut is raked from its own Meta day onto UTC; our own columns were already there.

ccspendshareimpressionsCPMregistrations (ours)cost/regwithin-daywROASpayers
US$155.5731.8%3,395$45.8258$2.682$113.810.7323
IN$78.0816.0%10,068$7.76142$0.550$63.970.8216
MX$21.784.5%5,231$4.1647$0.463$12.510.5713
ID$17.683.6%3,349$5.2843$0.411$42.512.405
GB$13.902.8%1,072$12.9620$0.695$73.505.297
PH$11.522.4%7,136$1.6143$0.268$20.251.766
TR$10.382.1%852$12.1719$0.546$43.764.224
AU$9.662.0%464$20.827$1.379$31.573.275
MY$8.881.8%1,947$4.5614$0.635$10.761.213
CL$8.291.7%566$14.676$1.382$0.000.001
UA$7.151.5%581$12.327$1.022$50.067.001
DE$7.131.5%246$28.943$2.375$7.111.006

184 further countries are not listed, $138.79 between them (28.4% of this block).

Reconciliation, what is measured and what is fitted. Every per-country cell is an allocation across hours; every total is measured. B1 (UTC-7) $106.33 from Meta 2026-08-18 and $293.62 from Meta 2026-08-19, a measured UTC-day total of $399.95; KBM1 (UTC-7) $19.98 from Meta 2026-08-18 and $43.05 from Meta 2026-08-19, a measured UTC-day total of $63.03; HR1 (UTC+0) needed none, its Meta day IS the UTC day at $25.83. The 1 UTC+0 account above is the control: raking one returns its country export unchanged, and that identity held here. Meta days this run did not pull were taken from the country archive at ad_ops/_data/country_archive/ — HR1 Meta 2026-08-19; KBM1 Meta 2026-08-18, 2026-08-19 — because a UTC-7 day reaches back into the Meta day before it and the run window did not.

(ii) The same cut per account, 2026-08-19.

B1

ccspendshareimpressionsCPMregistrations (ours)cost/regwithin-daywROASpayers
IN$78.0819.5%10,068$7.76141$0.554$63.970.8216
US$66.7116.7%2,771$24.0743$1.551$55.860.8417
MX$21.785.4%5,231$4.1646$0.473$12.510.5712
ID$17.684.4%3,349$5.2838$0.465$42.512.404
GB$13.903.5%1,072$12.9620$0.695$73.505.297
PH$11.522.9%7,136$1.6141$0.281$20.251.766
TR$10.382.6%852$12.1717$0.610$24.822.391
AU$9.662.4%464$20.827$1.379$31.573.274
MY$8.882.2%1,947$4.5614$0.635$10.761.212
CL$8.292.1%566$14.676$1.382$0.000.000
UA$7.151.8%581$12.327$1.022$50.067.001
DE$7.131.8%246$28.943$2.375$7.111.006

183 further countries are not listed, $138.79 between them (34.7% of this block).

Nomad Node

ccspendshareimpressionsCPMregistrations (ours)cost/regwithin-daywROASpayers
AR$0.000.0%0—2$0.000$0.00—0
AU$0.000.0%0—0—$0.00—1
AZ$0.000.0%0—0—$0.00—1
BD$0.000.0%0—1$0.000$0.00—0
BR$0.000.0%0—3$0.000$0.00—0
CL$0.000.0%0—0—$0.00—1
ES$0.000.0%0—1$0.000$0.00—0
HU$0.000.0%0—1$0.000$0.00—0
ID$0.000.0%0—5$0.000$0.00—1
IN$0.000.0%0—1$0.000$0.00—0
JO$0.000.0%0—1$0.000$0.00—0
KR$0.000.0%0—1$0.000$0.00—0

15 further countries are not listed, $0.00 between them (0.0% of this block).

KBM1

ccspendshareimpressionsCPMregistrations (ours)cost/regwithin-daywROASpayers
US$63.03100.0%358$176.0611$5.730$52.960.843

HR1

ccspendshareimpressionsCPMregistrations (ours)cost/regwithin-daywROASpayers
US$25.83100.0%266$97.114$6.457$4.990.191

Aggregate, both days

Pooled over the 24 hours both days delivered in, on the UTC clock. Meta's instrument on both sides of every ratio.

spendimpressionsCPMclicksCTRinstallsclick→installCPI
2026-08-19$399.9554,3307.363,3456.16%42812.80%$0.934
2026-08-20$425.8150,9678.353,1946.27%41913.12%$1.016
  • click → install 12.80% → 13.12% (+2.5%, p = 0.697) — did not separate.
  • CTR 6.16% → 6.27% (+1.8%, p = 0.460) — did not separate.

Pooling across hours is a mix comparison: it weights each hour by delivery, so these rows move where the hourly pairing does not.

2026-08-21T18:23:58.594732 image/svg+xml Matplotlib v3.11.1, https://matplotlib.org/ booked within-day cohort @ 24h 0 250 500 750 1000 1250 1500 1750 USD 3.59 1.28 1.95 4.21 0.70 0.83 the number above each bar is that measure's ROAS against the same day's spend The three revenue measures, 2026-08-19 against 2026-08-20 (UTC) 2026-08-19 2026-08-20
The three revenue measures on both days, ROAS above each bar
2026-08-21T18:23:58.641988 image/svg+xml Matplotlib v3.11.1, https://matplotlib.org/ 2026-08-19 2026-08-20 0 250 500 750 1000 1250 1500 1750 USD the bottom segment IS the within-day figure: the same number seen twice Booked revenue by the day its payer installed (UTC) the day itself one day earlier two days earlier three days earlier four or more
Booked revenue by the day its payer installed

The three measures disagree about the day: booked rose to $2,079.32 while within-day fell 1.17 to 0.56, because booked counts yesterday's buyers too.

And the two frozen measures disagree about the size of the fall: 46.5% sealed against 56.6% hour-neutral, and the gap between them is the censoring.

The instrument gate reads 1.18 on the META clock, against a 0.91–1.06 band. It bounds any comparison against Meta's revenue column and leaves within-day untouched.

The campaigns

Rank on CPM, cost per click and CPI only — geography is the treatment across these campaigns, and revenue still needs 16 payers.

⚠ Spend here is each account's OWN day, while the account tables above restitch it onto UTC. That is a seven-hour window shift on B1 and KBM1 and nothing at all on the two UTC+0 accounts, so a row here and the same account's row above are two different 24-hour windows.

(i) Every campaign the portfolio ran, 2026-08-20.

campaignaccountspendshareimpressionsCPMregistrationscost/regwithin-daywROASpayers
bailingxia_meituan_ww_cvr_260802B1$340.8282.6%37,196$9.16503$0.678$297.590.87121
bailingxia_srfhjyd_ww_cvr_260814KBM1$51.2812.4%255$201.1019$2.699$15.960.313
0814_vl_bailingxia_srfhjyd_maimaiNomad Node$20.304.9%2,270$8.9436$0.564$0.000.001
bailingxia_srfhjyd_ww_cvr_260819HR1$0.000.0%0—40$0.000$5.76—1
bailingxia_meituan_ww_cvr_260730B1$0.000.0%0—39$0.000$0.00—4
bailingxia_meituan_t3_cvr_260725B1$0.000.0%0—17$0.000$0.00—3
bailingxia_srfhjyd_ww_cvr_260816HR1$0.000.0%0—7$0.000$0.00—0
0811_vl_bailingxia_srfhjyd-rain-chuvaNomad Node$0.000.0%0—3$0.000$0.00—9
0811_vl_bailingxia_srfhjyd-healthNomad Node$0.000.0%0—1$0.000$0.00—0
bailingxia_meituan_t1_cvr_260725B1$0.000.0%0—1$0.000$0.00—0
0814_vl_bailingxia_srfhjyd-rain-chuvaNomad Node$0.000.0%0—1$0.000$0.00—5
0818_vl_bailingxia_srfhjyd_maimaiNomad Node$0.000.0%0—0—$0.00—1
bailingxia_meituan_dach_cvr_260729B1$0.000.0%0—0—$0.00—3

(ii) The same, per account, 2026-08-20.

B1

campaignaccountspendshareimpressionsCPMregistrationscost/regwithin-daywROASpayers
bailingxia_meituan_ww_cvr_260802B1$340.82100.0%37,196$9.16503$0.678$297.590.87121
bailingxia_meituan_ww_cvr_260730B1$0.000.0%0—39$0.000$0.00—4
bailingxia_meituan_t3_cvr_260725B1$0.000.0%0—17$0.000$0.00—3
bailingxia_meituan_t1_cvr_260725B1$0.000.0%0—1$0.000$0.00—0
bailingxia_meituan_dach_cvr_260729B1$0.000.0%0—0—$0.00—3

Nomad Node

campaignaccountspendshareimpressionsCPMregistrationscost/regwithin-daywROASpayers
0814_vl_bailingxia_srfhjyd_maimaiNomad Node$20.30100.0%2,270$8.9436$0.564$0.000.001
0811_vl_bailingxia_srfhjyd-rain-chuvaNomad Node$0.000.0%0—3$0.000$0.00—9
0811_vl_bailingxia_srfhjyd-healthNomad Node$0.000.0%0—1$0.000$0.00—0
0814_vl_bailingxia_srfhjyd-rain-chuvaNomad Node$0.000.0%0—1$0.000$0.00—5
0818_vl_bailingxia_srfhjyd_maimaiNomad Node$0.000.0%0—0—$0.00—1

KBM1

campaignaccountspendshareimpressionsCPMregistrationscost/regwithin-daywROASpayers
bailingxia_srfhjyd_ww_cvr_260814KBM1$51.28100.0%255$201.1019$2.699$15.960.313

HR1

campaignaccountspendshareimpressionsCPMregistrationscost/regwithin-daywROASpayers
bailingxia_srfhjyd_ww_cvr_260819HR1$0.000.0%0—40$0.000$5.76—1
bailingxia_srfhjyd_ww_cvr_260816HR1$0.000.0%0—7$0.000$0.00—0

⚠ These accounts run more than one campaign and their export carries no campaign column, so the rows above understate them: HR1. Build the column preset on that account itself.

This section carries no chart, and the gap is a standing one in the chart set.

Every asset

(i) Every asset the portfolio ran, 2026-08-20. One row per ad, the account named on it.

assetaccountspendimpressionsCPMregistrationscost/regwithin-daywROAScohort@24hpayers
vid_ww_pool5_15s_9x16_v1B1$340.8237,196$9.16460$0.741$294.190.86$349.9590
vid_ww_pool5_15s_9x16_v2KBM1$51.28255$201.1019$2.699$15.960.31$15.963
vid_ww_bodysuit_10s_9x16_v3HR1$22.571,262$17.8835$0.645$5.760.26$5.761
vid_ww_bodysuit_10s_9x16_v2HR1$20.34210$96.867$2.906$0.000.00$0.000
nomad-0814-03-maimaiNomad Node$20.302,270$8.9436$0.564$0.000.00$0.001
vid_ww_pool5_15s_9x16_v3HR1$6.56239$27.455$1.312$0.000.00$0.000
vid_ww_pool5_15s_9x16_v2HR1$0.544$135.000—$0.000.00$0.000
vid_ww_bodysuit_10s_9x16_v1B1$0.000—40$0.000$0.00—$0.0013
vid_ww_bikini_10s_9x16_v1B1$0.000—17$0.000$0.00—$0.005
vid_ww_lounge_8s_9x16_v1B1$0.000—11$0.000$3.40—$3.4013
vid_ww_spill_8s_9x16_v1 - CopyB1$0.000—11$0.000$0.00—$0.004
vid_sasian_bikini_6s_9x16_v1B1$0.000—6$0.000$0.00—$0.002
vid_sasian_bikini_10s_9x16_v1B1$0.000—3$0.000$0.00—$0.000
vid_sasian_spice5_15s_9x16_v1B1$0.000—3$0.000$0.00—$0.000
vid_utility_6s_9x16_v1B1$0.000—3$0.000$0.00—$0.000
nomad-0811-03-rain-chuva-video-01Nomad Node$0.000—2$0.000$0.00—$0.009
vid_ww_mugshot_15s_9x16_v1B1$0.000—2$0.000$0.00—$0.000
nomad-0814-03-rain-chuva-video-01Nomad Node$0.000—1$0.000$0.00—$0.005
vid_sasian_fishnet_top_6s_9x16_v1B1$0.000—1$0.000$0.00—$0.000
vid_utility_15s_9x16_v1B1$0.000—1$0.000$0.00—$0.000
vid_control_15s_9x16_v1B1$0.000—1$0.000$0.00—$0.000
vid_ww_spill_8s_9x16_v1B1$0.000—1$0.000$0.00—$0.000
nomad-0811-03-rain-chuva-02Nomad Node$0.000—1$0.000$0.00—$0.000
nomad-0811-04Nomad Node$0.000—1$0.000$0.00—$0.000
vid_sasian_bodysuit_10s_9x16_v1B1$0.000—0—$0.00—$0.001
vid_ww_bikini_10s_9x16_v1 - CopyB1$0.000—0—$0.00—$0.001
vid_ww_bodysuit_10s_9x16_v1 - CopyB1$0.000—0—$0.00—$0.001
nomad-0818-03-maimaiNomad Node$0.000—0—$0.00—$0.001
vid_de_german_mature_10s_9x16_v1B1$0.000—0—$0.00—$0.001

(ii) The same, per account, 2026-08-20.

B1

assetaccountspendimpressionsCPMregistrationscost/regwithin-daywROAScohort@24hpayers
vid_ww_pool5_15s_9x16_v1B1$340.8237,196$9.16460$0.741$294.190.86$349.9590
vid_ww_bodysuit_10s_9x16_v1B1$0.000—40$0.000$0.00—$0.0013
vid_ww_bikini_10s_9x16_v1B1$0.000—17$0.000$0.00—$0.005
vid_ww_spill_8s_9x16_v1 - CopyB1$0.000—11$0.000$0.00—$0.004
vid_ww_lounge_8s_9x16_v1B1$0.000—11$0.000$3.40—$3.4013
vid_sasian_bikini_6s_9x16_v1B1$0.000—6$0.000$0.00—$0.002
vid_utility_6s_9x16_v1B1$0.000—3$0.000$0.00—$0.000
vid_sasian_bikini_10s_9x16_v1B1$0.000—3$0.000$0.00—$0.000
vid_sasian_spice5_15s_9x16_v1B1$0.000—3$0.000$0.00—$0.000
vid_ww_mugshot_15s_9x16_v1B1$0.000—2$0.000$0.00—$0.000
vid_utility_15s_9x16_v1B1$0.000—1$0.000$0.00—$0.000
vid_control_15s_9x16_v1B1$0.000—1$0.000$0.00—$0.000
vid_ww_spill_8s_9x16_v1B1$0.000—1$0.000$0.00—$0.000
vid_sasian_fishnet_top_6s_9x16_v1B1$0.000—1$0.000$0.00—$0.000
vid_sasian_bodysuit_10s_9x16_v1B1$0.000—0—$0.00—$0.001
vid_ww_bodysuit_10s_9x16_v1 - CopyB1$0.000—0—$0.00—$0.001
vid_de_german_mature_10s_9x16_v1B1$0.000—0—$0.00—$0.001
vid_ww_bikini_10s_9x16_v1 - CopyB1$0.000—0—$0.00—$0.001

Nomad Node

assetaccountspendimpressionsCPMregistrationscost/regwithin-daywROAScohort@24hpayers
nomad-0814-03-maimaiNomad Node$20.302,270$8.9436$0.564$0.000.00$0.001
nomad-0811-03-rain-chuva-video-01Nomad Node$0.000—2$0.000$0.00—$0.009
nomad-0811-04Nomad Node$0.000—1$0.000$0.00—$0.000
nomad-0811-03-rain-chuva-02Nomad Node$0.000—1$0.000$0.00—$0.000
nomad-0814-03-rain-chuva-video-01Nomad Node$0.000—1$0.000$0.00—$0.005
nomad-0818-03-maimaiNomad Node$0.000—0—$0.00—$0.001

KBM1

assetaccountspendimpressionsCPMregistrationscost/regwithin-daywROAScohort@24hpayers
vid_ww_pool5_15s_9x16_v2KBM1$51.28255$201.1019$2.699$15.960.31$15.963

HR1

assetaccountspendimpressionsCPMregistrationscost/regwithin-daywROAScohort@24hpayers
vid_ww_bodysuit_10s_9x16_v3HR1$22.571,262$17.8835$0.645$5.760.26$5.761
vid_ww_bodysuit_10s_9x16_v2HR1$20.34210$96.867$2.906$0.000.00$0.000
vid_ww_pool5_15s_9x16_v3HR1$6.56239$27.455$1.312$0.000.00$0.000
vid_ww_pool5_15s_9x16_v2HR1$0.544$135.000—$0.000.00$0.000

The standing per-asset tables

#### A. Per asset — delivery

Meta's instrument on both sides of every ratio. Our registrations never enter this table.

assetspendimpressionsCPMclicksCTRinstallsclick→installCPI
E2/pool5$425.8150,9678.353,1946.27%41913.12%$1.016
earner/bodysuit$0.000—0—0——
earner/bikini$0.000—0—0——
earner/pool5$0.000—0—0——
E2/bikini$0.000—0—0——
E2/bodysuit$0.000—0—0——
E2/mugshot$0.000—0—0——
E2/spill$0.000—0—0——
E2/spill2$0.000—0—0——
E2/lounge$0.000—0—0——

#### B. Per asset — the three revenue measures

Ours on every column. Meta can express only booked, so it rides bracketed beside that column and never as a row. Payers are distinct people, never transactions. Cohort horizon H=24h, 100% covered on this day. Report clock UTC.

assetbookedbROAS (Meta)within-daywROAScohort@HcROASpayers (d0)payer% (d0)ARPU (d0)ARPPU (d0)largest payer
E2/pool5$1,122.212.64 (3.31)$294.190.69$349.950.82378.2%0.7739.4611.9%
earner/bodysuit$18.57— (—)$0.00—$0.00—00.0%0.000——
earner/bikini$19.68— (—)$0.00—$0.00—00.0%0.000——
earner/pool5$0.00— (—)$0.00—$0.00—00.0%0.000——
E2/bikini$12.94— (—)$0.00—$0.00—00.0%0.000——
E2/bodysuit$203.59— (—)$0.00—$0.00—00.0%0.000——
E2/mugshot$0.00— (—)$0.00—$0.00—00.0%0.000——
E2/spill$0.00— (—)$0.00—$0.00—00.0%0.000——
E2/spill2$173.26— (—)$0.00—$0.00—00.0%0.000——
E2/lounge$194.04— (—)$3.40—$3.40—19.1%0.3093.40100.0%

The 16-payer bar (#37): any row above whose payer count is under 16 is a direction and never a magnitude.

#### C. Per asset — country mix

⚠ This table is cut on the META day, while every other table on this page is on the UTC clock. Meta serves no hour × country grid, so per-country delivery exists only on the ad account's own UTC-7 day and cannot be restitched. It is a seven-hour-offset window against tables A and B above; read it for mix, never for level against them.

The last column is the share of that asset's own within-day revenue that came from the United States: where it far exceeds the US spend share, the return is a property of the geography the asset bought.

assetspendIndia %US %other %US % of its revenue
E2/pool5$340.8221.1%14.1%64.7%40.0%
2026-08-21T18:23:58.689983 image/svg+xml Matplotlib v3.11.1, https://matplotlib.org/ 0 2 4 6 8 10 12 14 16 18 20 22 hour (UTC) 5 6 7 8 9 10 CTR % CTR by hour — every asset, 2026-08-20 (UTC) E2/pool5
CTR by hour, one line per asset
2026-08-21T18:23:58.783921 image/svg+xml Matplotlib v3.11.1, https://matplotlib.org/ 0 2 4 6 8 10 12 14 16 18 20 22 hour (UTC) 6 8 10 12 14 16 18 20 22 installs / clicks % Click → install by hour — every asset, 2026-08-20 (UTC) E2/pool5
Click-to-install by hour, one line per asset
2026-08-21T18:23:58.827998 image/svg+xml Matplotlib v3.11.1, https://matplotlib.org/ 0 2 4 6 8 10 12 14 16 18 20 22 hour (UTC) 0.8 1.0 1.2 1.4 1.6 USD Cost per install by hour — every asset, 2026-08-20 (UTC) E2/pool5
Cost per install by hour, one line per asset
2026-08-21T18:23:58.735988 image/svg+xml Matplotlib v3.11.1, https://matplotlib.org/ 0 2 4 6 8 10 12 14 16 18 20 22 hour (UTC) 7 8 9 10 11 12 13 14 USD per 1,000 impressions CPM by hour — every asset, 2026-08-20 (UTC) E2/pool5
CPM by hour, one line per asset
2026-08-21T18:23:58.918357 image/svg+xml Matplotlib v3.11.1, https://matplotlib.org/ E2/pool5 0 20 40 60 80 100 % of that asset's spend Where each asset bought — share of its own spend by region, 2026-08-20 (META day) IN US GB ID MX TR other
Where each asset bought: share of its own spend by region
2026-08-21T18:23:58.960371 image/svg+xml Matplotlib v3.11.1, https://matplotlib.org/ E2/pool5 0.0 0.5 1.0 1.5 2.0 USD per install Cost per install by region — every asset, 2026-08-20 (META day) IN US GB ID MX TR
Cost per install by region, every asset
2026-08-21T18:23:59.015009 image/svg+xml Matplotlib v3.11.1, https://matplotlib.org/ E2/pool5 0 20 40 60 80 100 120 USD Within-day revenue by region — every asset, 2026-08-20 (META day) IN US GB ID MX
Within-day revenue by region, every asset
2026-08-21T18:23:59.082185 image/svg+xml Matplotlib v3.11.1, https://matplotlib.org/ 08-17 08-18 08-19 08-20 day (UTC) 0 2 4 6 8 payers / installs % Payer rate by day — every asset (UTC) E2/pool5 earner/bodysuit E2/bikini E2/bodysuit
Payer rate by day, one line per asset
2026-08-21T18:23:59.127865 image/svg+xml Matplotlib v3.11.1, https://matplotlib.org/ 08-17 08-18 08-19 08-20 day (UTC) 0.0 0.2 0.4 0.6 0.8 1.0 within-day USD per install ARPU by day — every asset (UTC) E2/pool5 earner/bodysuit E2/bikini E2/bodysuit
ARPU by day, one line per asset
2026-08-21T18:23:59.174464 image/svg+xml Matplotlib v3.11.1, https://matplotlib.org/ 08-17 08-18 08-19 08-20 day (UTC) 0 2 4 6 8 10 12 14 within-day USD per payer ARPPU by day — every asset (UTC) E2/pool5 earner/bodysuit E2/bikini E2/bodysuit
ARPPU by day, one line per asset
2026-08-21T18:23:59.213688 image/svg+xml Matplotlib v3.11.1, https://matplotlib.org/ 08-17 08-18 08-19 08-20 day (UTC) 0.7 0.8 0.9 1.0 1.1 1.2 revenue / spend Within-day ROAS by day — every asset (UTC) E2/pool5
Within-day ROAS by day, one line per asset
2026-08-21T18:23:59.266585 image/svg+xml Matplotlib v3.11.1, https://matplotlib.org/ CTR click→install CPI payer rate * ARPU * wROAS * 0.0 0.2 0.4 0.6 0.8 1.0 normalised across the day (1.0 = best of the pack) * axis sits under the 16-payer bar on most days: direction only, never a level. Assets under 20 installs on the day are not plotted. Composite — every asset across every measure, 2026-08-20 E2/pool5
Composite: every asset across every measure, normalised across the day

Sinai — every section below this line is Sinai's: the day, the hours, the countries, the campaigns and the assets, counted on Sinai's own ledger and its own accounts. Use the buttons above to switch.

The day, and every account

(i) Sinai, its first day, with the one-day United States test and without it. Every other table in this tab is the without-it column.

with the US testwithout iteffect of excluding it
spend$59.07$33.16−43.9%
impressions1,8601,712−8.0%
CPM$31.76$19.37−39.0%
registrations (ours)122117−4.1%
cost per registration$0.484$0.283−41.5%
booked$74.13 (1.26)$68.15 (2.06)−8.1%
within-day$74.13 (1.26)$68.15 (2.06)−8.1%
cohort @ 24h$119.62 (2.03)$113.64 (3.43)−5.0%
booked payers1412−2
payer rate11.5%10.3%−1.2 pts
ARPU$0.608$0.583−4.1%
ARPPU$5.30$5.68+7.3%

(ii) Sinai, its first day, every account. There is no prior day.

2026-08-20 (the day) — delivery

accountspendshareimpressionsCPMregistrationscost/reg
Portfolio$33.16100%1,712$19.37117$0.283
B1$13.8942%621$22.3750$0.278
Nomad Node$0.000%0—0—
KBM1$0.000%0—0—
HR1$19.2758%1,091$17.6667$0.288

2026-08-20 (the day) — the three measures

accountbookedbROASwithin-daywROAScohort@24hcROASpayerspayer%ARPUARPPU
Portfolio$68.152.06$68.152.06$113.643.431210.3%$0.583$5.68
B1$26.281.89$26.281.89$45.893.30510.0%$0.526$5.26
Nomad Node$0.00—$0.00—$0.00—0———
KBM1$0.00—$0.00—$0.00—0———
HR1$41.872.17$41.872.17$67.753.52710.4%$0.625$5.98

Each account, 2026-08-20. There is no prior day to compare against: nothing here delivered before this day.

B1

2026-08-20
spend$13.89
impressions621
CPM$22.37
registrations (ours)50
cost per registration$0.278
booked$26.28 (1.89)
within-day$26.28 (1.89)
cohort @ 24h$45.89 (3.30)
booked payers5
payer rate (d0)10.0%
ARPU (d0)$0.9178
ARPPU (d0)$9.18

Nomad Node — nothing delivered and nobody registered on this day.

KBM1 — nothing delivered and nobody registered on this day.

HR1

2026-08-20
spend$19.27
impressions1,091
CPM$17.66
registrations (ours)67
cost per registration$0.288
booked$41.87 (2.17)
within-day$41.87 (2.17)
cohort @ 24h$67.75 (3.52)
booked payers7
payer rate (d0)13.4%
ARPU (d0)$1.0111
ARPPU (d0)$7.53

Hourly, 2026-08-20

Sinai's hourly cut is below the floor to read. 1,712 impressions put nearly every hour under the 50-impression minimum, and payers run 0–3 an hour.

(i) The portfolio, hour by hour, 2026-08-20. Every account placeable on the report clock, summed; delivery on the delivery hour, our columns on the hour the appendix names.

hrspendimpressionsCPMregistrationscost/regbookedbROASwithin-daywROAScohort@24hcROASbooked payers
00$0.000—0—$0.00—$0.00—$0.00—0
01$0.000—0—$0.00—$0.00—$0.00—0
02$0.000—0—$0.00—$0.00—$0.00—0
03$0.000—0—$0.00—$0.00—$0.00—0
04$0.000—0—$0.00—$0.00—$0.00—0
05$0.000—0—$0.00—$0.00—$0.00—0
06$0.000—0—$0.00—$0.00—$0.00—0
07$0.000—0—$0.00—$0.00—$0.00—0
08$0.000—0—$0.00—$0.00—$0.00—0
09$0.6645$14.673$0.220$0.000.00$0.000.00$0.000.000
10$2.19123$17.808$0.274$0.000.00$3.291.50$9.874.510
11$1.3198$13.377$0.187$0.000.00$11.959.12$11.959.120
12$4.00160$25.003$1.333$2.990.75$0.000.00$3.350.841
13$2.32116$20.002$1.160$0.000.00$0.000.00$0.000.000
14$0.8759$14.753$0.290$0.000.00$0.000.00$0.000.000
15$1.3187$15.064$0.328$7.125.43$7.125.43$7.125.432
16$0.000—1$0.000$0.00—$0.00—$0.00—0
17$0.000—0—$0.00—$0.00—$0.00—0
18$0.000—0—$0.00—$0.00—$0.00—0
19$2.40145$16.556$0.400$6.552.73$13.675.70$24.3510.142
20$4.03201$20.0517$0.237$12.533.11$2.990.74$2.990.743
21$2.74117$23.4215$0.183$10.223.73$3.671.34$3.671.343
22$6.08264$23.0321$0.290$5.950.98$9.511.56$18.443.031
23$5.25297$17.6827$0.194$22.804.34$15.953.04$31.906.084

Absent from the table above, by name: Nomad Node. Left out only when it cannot be placed on the report clock; its money is still in every day-grain table.

(ii) The same 24 hours per account, 2026-08-20.

B1 — $13.89 on this day

hrspendimpressionsCPMregistrationscost/regbookedbROASwithin-daywROAScohort@24hcROASbooked payers
00$0.000—0—$0.00—$0.00—$0.00—0
01$0.000—0—$0.00—$0.00—$0.00—0
02$0.000—0—$0.00—$0.00—$0.00—0
03$0.000—0—$0.00—$0.00—$0.00—0
04$0.000—0—$0.00—$0.00—$0.00—0
05$0.000—0—$0.00—$0.00—$0.00—0
06$0.000—0—$0.00—$0.00—$0.00—0
07$0.000—0—$0.00—$0.00—$0.00—0
08$0.000—0—$0.00—$0.00—$0.00—0
09$0.000—0—$0.00—$0.00—$0.00—0
10$0.000—0—$0.00—$0.00—$0.00—0
11$0.000—0—$0.00—$0.00—$0.00—0
12$0.000—0—$0.00—$0.00—$0.00—0
13$0.000—0—$0.00—$0.00—$0.00—0
14$0.000—0—$0.00—$0.00—$0.00—0
15$0.239$25.561$0.230$0.000.00$0.000.00$0.000.000
16$0.000—0—$0.00—$0.00—$0.00—0
17$0.000—0—$0.00—$0.00—$0.00—0
18$0.000—0—$0.00—$0.00—$0.00—0
19$2.40145$16.556$0.400$6.552.73$13.675.70$24.3510.142
20$4.03201$20.0517$0.237$6.551.63$2.990.74$2.990.742
21$2.74117$23.4215$0.183$7.232.64$3.671.34$3.671.342
22$2.5380$31.626$0.422$5.952.35$5.952.35$14.885.881
23$1.9669$28.415$0.392$0.000.00$0.000.00$0.000.000

KBM1 — $0.00 on this day

hrspendimpressionsCPMregistrationscost/regbookedbROASwithin-daywROAScohort@24hcROASbooked payers
00$0.000—0—$0.00—$0.00—$0.00—0
01$0.000—0—$0.00—$0.00—$0.00—0
02$0.000—0—$0.00—$0.00—$0.00—0
03$0.000—0—$0.00—$0.00—$0.00—0
04$0.000—0—$0.00—$0.00—$0.00—0
05$0.000—0—$0.00—$0.00—$0.00—0
06$0.000—0—$0.00—$0.00—$0.00—0
07$0.000—0—$0.00—$0.00—$0.00—0
08$0.000—0—$0.00—$0.00—$0.00—0
09$0.000—0—$0.00—$0.00—$0.00—0
10$0.000—0—$0.00—$0.00—$0.00—0
11$0.000—0—$0.00—$0.00—$0.00—0
12$0.000—0—$0.00—$0.00—$0.00—0
13$0.000—0—$0.00—$0.00—$0.00—0
14$0.000—0—$0.00—$0.00—$0.00—0
15$0.000—0—$0.00—$0.00—$0.00—0
16$0.000—0—$0.00—$0.00—$0.00—0
17$0.000—0—$0.00—$0.00—$0.00—0
18$0.000—0—$0.00—$0.00—$0.00—0
19$0.000—0—$0.00—$0.00—$0.00—0
20$0.000—0—$0.00—$0.00—$0.00—0
21$0.000—0—$0.00—$0.00—$0.00—0
22$0.000—0—$0.00—$0.00—$0.00—0
23$0.000—0—$0.00—$0.00—$0.00—0

HR1 — $19.27 on this day

hrspendimpressionsCPMregistrationscost/regbookedbROASwithin-daywROAScohort@24hcROASbooked payers
00$0.000—0—$0.00—$0.00—$0.00—0
01$0.000—0—$0.00—$0.00—$0.00—0
02$0.000—0—$0.00—$0.00—$0.00—0
03$0.000—0—$0.00—$0.00—$0.00—0
04$0.000—0—$0.00—$0.00—$0.00—0
05$0.000—0—$0.00—$0.00—$0.00—0
06$0.000—0—$0.00—$0.00—$0.00—0
07$0.000—0—$0.00—$0.00—$0.00—0
08$0.000—0—$0.00—$0.00—$0.00—0
09$0.6645$14.673$0.220$0.000.00$0.000.00$0.000.000
10$2.19123$17.808$0.274$0.000.00$3.291.50$9.874.510
11$1.3198$13.377$0.187$0.000.00$11.959.12$11.959.120
12$4.00160$25.003$1.333$2.990.75$0.000.00$3.350.841
13$2.32116$20.002$1.160$0.000.00$0.000.00$0.000.000
14$0.8759$14.753$0.290$0.000.00$0.000.00$0.000.000
15$1.0878$13.853$0.360$7.126.59$7.126.59$7.126.592
16$0.000—1$0.000$0.00—$0.00—$0.00—0
17$0.000—0—$0.00—$0.00—$0.00—0
18$0.000—0—$0.00—$0.00—$0.00—0
19$0.000—0—$0.00—$0.00—$0.00—0
20$0.000—0—$5.98—$0.00—$0.00—1
21$0.000—0—$2.99—$0.00—$0.00—1
22$3.55184$19.2915$0.237$0.000.00$3.561.00$3.561.000
23$3.29228$14.4322$0.150$22.806.93$15.954.85$31.909.704

Nomad Node — no hourly block: it delivered nothing on this day. Its export returned No data available, a measured zero, and the zero row is in the day-grain tables.

Country, 2026-08-20

2026-08-23T13:52:37.119682 image/svg+xml Matplotlib v3.11.1, https://matplotlib.org/ IN TR US BR ID UA CL AR CO MX IQ AU 0.0 2.5 5.0 7.5 10.0 12.5 15.0 17.5 20.0 USD delivery raked onto the report clock against two measured margins; revenue is the report day Spend against within-day revenue by country — 2026-08-20 spend within-day revenue
Spend against within-day revenue, by country
2026-08-23T13:52:37.157102 image/svg+xml Matplotlib v3.11.1, https://matplotlib.org/ −0.04 −0.02 0.00 0.02 0.04 USD per install Cost per install by country — 2026-08-20 (report clock)
Cost per install by country

(i) The portfolio by country, 2026-08-20. On the report clock. Meta serves no hour x country grid, so each country cut is raked from its own Meta day onto UTC; our own columns were already there.

ccspendshareimpressionsCPMregistrations (ours)cost/regwithin-daywROASpayers
IN$9.3028.1%649$14.3435$0.266$21.362.304
TR$2.517.6%89$28.319$0.279$0.000.000
US$2.347.0%25$94.442$1.168$0.000.000
BR$1.765.3%91$19.3410$0.176$0.000.000
ID$1.504.5%116$12.895$0.300$0.000.000
UA$1.033.1%11$93.321$1.026$0.000.000
CL$0.812.4%32$25.085$0.161$0.000.000
AR$0.792.4%63$12.605$0.159$0.000.000
CO$0.762.3%31$24.233$0.253$0.000.000
MX$0.742.2%45$16.562$0.371$0.000.000
IQ$0.702.1%31$22.152$0.348$5.958.561
AU$0.682.1%8$80.740—$0.000.000

93 further countries are not listed, $10.24 between them (30.9% of this block).

Reconciliation, what is measured and what is fitted. Every per-country cell is an allocation across hours; every total is measured. B1 (UTC-7) $13.89 from Meta 2026-08-20, a measured UTC-day total of $13.89; HR1 (UTC+0) needed none, its Meta day IS the UTC day at $19.27. The 1 UTC+0 account above is the control: raking one returns its country export unchanged, and that identity held here.

Absent from the table above, by name: Nomad Node, KBM1. Their spend is in every day-grain table; only their country split is missing.

(ii) The same cut per account, 2026-08-20.

B1

ccspendshareimpressionsCPMregistrations (ours)cost/regwithin-daywROASpayers
IN$3.8827.9%274$14.1914$0.277$10.682.751
US$2.0714.9%22$95.052$1.033$0.000.000
TR$0.997.2%32$31.275$0.199$0.000.000
ID$0.634.5%24$25.802$0.315$0.000.000
CL$0.594.2%19$30.623$0.195$0.000.000
CO$0.392.8%15$25.391$0.389$0.000.000
MX$0.352.5%27$13.141$0.352$0.000.000
AR$0.342.5%21$16.373$0.114$0.000.000
IT$0.312.3%8$37.741$0.314$3.6711.681
IQ$0.312.2%14$21.222$0.153$5.9519.501
UA$0.302.1%4$74.090—$0.000.000
CA$0.282.0%7$41.932$0.139$0.000.000

79 further countries are not listed, $3.45 between them (24.9% of this block).

HR1

ccspendshareimpressionsCPMregistrations (ours)cost/regwithin-daywROASpayers
IN$5.4228.1%375$14.4521$0.258$10.681.973
BR$1.769.1%91$19.3410$0.176$0.000.000
TR$1.527.9%57$26.674$0.380$0.000.000
ID$0.874.5%92$9.463$0.290$0.000.000
UA$0.733.8%7$104.291$0.730$0.000.000
AU$0.502.6%5$100.000—$0.000.000
AR$0.452.3%42$10.712$0.225$0.000.000
IQ$0.392.0%17$22.940—$0.000.000
MX$0.392.0%18$21.671$0.390$0.000.000
CA$0.371.9%3$123.332$0.185$11.9532.311
CO$0.371.9%16$23.122$0.185$0.000.000
KR$0.371.9%11$33.641$0.370$0.000.000

66 further countries are not listed, $6.13 between them (31.8% of this block).

Nomad Node — no country block: it delivered nothing on this day, a measured zero.

KBM1 — no country block: it delivered nothing on this day, a measured zero.

The prior day, 2026-08-19, the same two tables.

(i) The portfolio by country, 2026-08-19. On the report clock. Meta serves no hour x country grid, so each country cut is raked from its own Meta day onto UTC; our own columns were already there.

ccspendshareimpressionsCPMregistrations (ours)cost/regwithin-daywROASpayers

Absent from the table above, by name: B1, Nomad Node, KBM1, HR1. Their spend is in every day-grain table; only their country split is missing.

(ii) The same cut per account, 2026-08-19.

B1 — no country block: it delivered nothing on this day, a measured zero.

Nomad Node — no country block: it delivered nothing on this day, a measured zero.

KBM1 — no country block: it delivered nothing on this day, a measured zero.

HR1 — no country block: it delivered nothing on this day, a measured zero.

Aggregate, both days

There is no aggregate comparison: Sinai first delivered on 2026-08-20. The pooled funnel, its tests and the booked-split chart arrive on the 08-21 page.

What this day settles is the horizon: all 117 registrations have lived 24 hours, so cohort@24h is frozen at 3.43 against a sealed 2.06.

The campaigns

⚠ Spend here is each account's OWN day, while the account tables above restitch it onto UTC. That is a seven-hour window shift on B1 and KBM1 and nothing at all on the two UTC+0 accounts, so a row here and the same account's row above are two different 24-hour windows. KBM1 is the visible case: its worldwide arm reads $10.04 here and $0.00 above, because every cent of it was spent in account hours 18 to 23 and belongs to UTC 08-21.

(i) Every campaign the portfolio ran, 2026-08-20.

campaignaccountspendshareimpressionsCPMregistrationscost/regwithin-daywROASpayers
bailingxia_srfhjyd_sn_ww_cvr_260820B1$39.0357.1%1,944$20.0850$0.781$26.280.675
bailingxia_srfhjyd_sn_ww_cvr_260820HR1$19.2728.2%1,091$17.6667$0.288$41.872.177
bailingxia_srfhjyd_sn_ww_cvr_260820KBM1$10.0414.7%906$11.080—$0.000.000

(ii) The same, per account, 2026-08-20.

B1

campaignaccountspendshareimpressionsCPMregistrationscost/regwithin-daywROASpayers
bailingxia_srfhjyd_sn_ww_cvr_260820B1$39.03100.0%1,944$20.0850$0.781$26.280.675

Nomad Node

nothing delivered or registered.

KBM1

campaignaccountspendshareimpressionsCPMregistrationscost/regwithin-daywROASpayers
bailingxia_srfhjyd_sn_ww_cvr_260820KBM1$10.04100.0%906$11.080—$0.000.000

HR1

campaignaccountspendshareimpressionsCPMregistrationscost/regwithin-daywROASpayers
bailingxia_srfhjyd_sn_ww_cvr_260820HR1$19.27100.0%1,091$17.6667$0.288$41.872.177

Every asset

(i) Every asset the portfolio ran, 2026-08-20. One row per ad, the account named on it.

assetaccountspendimpressionsCPMregistrationscost/regwithin-daywROAScohort@24hpayers
vid_ww_poolslow_15s_9x16_sn_v1B1$15.23757$20.120—$0.000.00$0.000
vid_ww_bodysuit_10s_9x16_sn_v1HR1$13.80764$18.0648$0.288$15.511.12$21.852
vid_ww_poolphoto_15s_9x16_sn_v1B1$11.67641$18.215$2.334$0.000.00$0.000
vid_ww_lace_10s_9x16_sn_v1B1$6.96268$25.975$1.392$5.950.86$14.881
vid_ww_pool5_15s_9x16_sn_v1HR1$5.47327$16.7319$0.288$26.364.82$45.905
vid_ww_poolslow_15s_9x16_sn_v1KBM1$4.42408$10.830—$0.000.00$0.000
vid_ww_shirt_10s_9x16_sn_v1B1$2.32125$18.562$1.160$0.000.00$0.000
vid_ww_kimono_10s_9x16_sn_v1B1$1.91118$16.192$0.955$0.000.00$0.000
vid_ww_lace_10s_9x16_sn_v1KBM1$1.6578$21.150—$0.000.00$0.000
vid_ww_shirt_10s_9x16_sn_v1KBM1$1.22162$7.530—$0.000.00$0.000
vid_ww_kimono_10s_9x16_sn_v1KBM1$1.20149$8.050—$0.000.00$0.000
vid_ww_pooldirect_15s_9x16_sn_v1KBM1$0.9962$15.970—$0.000.00$0.000
vid_ww_pooldirect_15s_9x16_sn_v1B1$0.9435$26.863$0.313$2.993.18$2.991
vid_ww_poolphoto_15s_9x16_sn_v1KBM1$0.5647$11.910—$0.000.00$0.000
vid_ww_pool5_15s_9x16_sn_v1B1$0.000—33$0.000$17.34—$28.023

(ii) The same, per account, 2026-08-20.

B1

assetaccountspendimpressionsCPMregistrationscost/regwithin-daywROAScohort@24hpayers
vid_ww_poolslow_15s_9x16_sn_v1B1$15.23757$20.120—$0.000.00$0.000
vid_ww_poolphoto_15s_9x16_sn_v1B1$11.67641$18.215$2.334$0.000.00$0.000
vid_ww_lace_10s_9x16_sn_v1B1$6.96268$25.975$1.392$5.950.86$14.881
vid_ww_shirt_10s_9x16_sn_v1B1$2.32125$18.562$1.160$0.000.00$0.000
vid_ww_kimono_10s_9x16_sn_v1B1$1.91118$16.192$0.955$0.000.00$0.000
vid_ww_pooldirect_15s_9x16_sn_v1B1$0.9435$26.863$0.313$2.993.18$2.991
vid_ww_pool5_15s_9x16_sn_v1B1$0.000—33$0.000$17.34—$28.023

Nomad Node

nothing delivered or registered.

KBM1

assetaccountspendimpressionsCPMregistrationscost/regwithin-daywROAScohort@24hpayers
vid_ww_poolslow_15s_9x16_sn_v1KBM1$4.42408$10.830—$0.000.00$0.000
vid_ww_lace_10s_9x16_sn_v1KBM1$1.6578$21.150—$0.000.00$0.000
vid_ww_shirt_10s_9x16_sn_v1KBM1$1.22162$7.530—$0.000.00$0.000
vid_ww_kimono_10s_9x16_sn_v1KBM1$1.20149$8.050—$0.000.00$0.000
vid_ww_pooldirect_15s_9x16_sn_v1KBM1$0.9962$15.970—$0.000.00$0.000
vid_ww_poolphoto_15s_9x16_sn_v1KBM1$0.5647$11.910—$0.000.00$0.000

HR1

assetaccountspendimpressionsCPMregistrationscost/regwithin-daywROAScohort@24hpayers
vid_ww_bodysuit_10s_9x16_sn_v1HR1$13.80764$18.0648$0.288$15.511.12$21.852
vid_ww_pool5_15s_9x16_sn_v1HR1$5.47327$16.7319$0.288$26.364.82$45.905

⚠ One B1 ad appears twice in Sinai's table, and it is one ad. 120251228633720364 was renamed from vid_ww_pool5_15s_9x16_sn_v1 to vid_ww_poolslow_15s_9x16_sn_v1 on 08-20. Meta reports its spend under the new name while every registration it produced carries the tracker string captured at click time, which holds the old one. The rows are the same object either side of a rename; the ad id is what joins them. This is the reason HR1's and KBM1's ads were deliberately left with names that no longer describe what they play.

3Caveats

Every cohort@24h figure on this page is frozen, and the page is a re-cut of the day. Payments were pulled at 2026-08-22 00:46 UTC for Vloom and 03:03 UTC for Sinai. The day's last registration landed at 23:59 UTC on 08-20 and passed its 24-hour horizon at 23:59 UTC on 08-21, so coverage is 100% on both 08-19 and 08-20 and neither column can rise again — for Sinai's buy that is 117 of 117 registrations matured. This page as first published on 2026-08-21 carried the same cohort at 34% coverage for Vloom and 0% for Sinai's buy, and declined every comparison on that row. Those were lower-coverage reads of the same cohort, not different numbers: within-day is unchanged, booked is unchanged, and the cohort columns have finished rising.

Within-day is censored by time of day, and two of this page's comparisons turn on it. It gives a 23:00 registration one hour to pay before the day closes and a 01:00 registration twenty-three, so it penalises whichever day bought later in the clock — and that was 08-19: 36.2% of 08-19's registrations landed in hours 18–23 against 26.7% of 08-20's. Removing the censoring lifts 08-19 by 44.6% and 08-20 by only 17.5%, which is the whole distance between a 46.5% sealed fall and a 56.6% hour-neutral one.

The two products cannot be compared on the sealed measure, because they started buying at different times of day. Sinai's installs sat a median 2.7 hours before midnight against Vloom's 11.7. Sinai began buying at 09:00 UTC and 74% of its registrations had under six hours left before midnight, and that is the whole $45.49 between its sealed $68.15 and the $113.64 it reached at the horizon. Cohort at a fixed horizon counts each person from their own install and does not depend on launch time.

Sinai's 12 payers are below the 16-payer bar, as are every per-account and per-asset payer count in its tables. Nothing in Sinai's tab is a magnitude — including the 3.43, which is the largest return on this page and rests on twelve people.

The one-day United States test is excluded from every Sinai figure on this page. bailingxia_srfhjyd_sn_us_cvr_260820 ran United States only in KBM1 for one day: $25.91 at a $175.07 CPM, 5 registrations at $5.182 each and $5.98 back from 2 payers, a 0.23 return. Pooled into the app it drags the return from 2.06 to 1.26 and the cost per registration from $0.283 to $0.484. Sinai's tab carries both scopes side by side on its first table; every other table there is without it.

Every Sinai payment now joins to a registration. All 113 pulled across 08-18 → 08-22 resolve to a person, 23 of 23 on 08-20 itself, eleven of them repaired through their account UUID where the API row carried no payer. At the cut this page first shipped on, 20 of 08-20's 23 joined. A payment that joins to nobody is dropped from all three revenue measures, booked included, which is why Sinai's sealed within-day reads $68.15 here where it read $65.16 then. Within-day's seal is on the clock, and the join sits outside it. The same 23 payments were on the ledger at both cuts, and three of them had no person to attach to.

HR1's bailingxia_srfhjyd_ww_cvr_260819 was counted in no account until this run. It was built on 08-19 and first delivered on Meta day 08-20, so no published page is wrong, but the unknown-id gate had it at 42 registration rows attributed to nothing. It is now enumerated. Reading that gate at the end of every run is what caught it.

A third instrument defect was found and fixed with them, and it moved a number on this page. A UTC day with no hourly rows for the app being reported kept its DAY-GRAIN figure, which on a UTC-7 account is a different 24 hours. KBM1's worldwide Sinai arm delivered $10.04 entirely in account hours 18–23, so all of it belongs to UTC 08-21 and this day is $0.00 — and the day grain had put the whole $10.04 into 08-20, overstating Sinai's spend by 30%. A zero is a result of the stitch and stands as the day's figure. Vloom's figures are unchanged by the fix.

Two further instrument defects were found and fixed while staging this page, and both would have misreported Sinai. A day-grain export was filtered on the ad-set substring alone, so an account holding a second app's campaigns summed both into one figure; and an account with no campaigns for the app being reported fell through to yielding every row, which put $568.55 of unrelated pixvoo and greenvid spend into a $59.07 day. Both are corrected in portfolio_split.py, and every Vloom figure on this page is byte-identical before and after the fix.

The Sinai registration feed is filtered on a field that was being discarded. The sink push has always carried 应用, either lum or sin, and the users file dropped it, so the two apps' registrations sat in one file distinguishable only by campaign id, which cannot separate their organic traffic at all. The field is now kept and --app filters on it.

Adjust is no longer the source for Sinai and is retained as a cross-check, and it now agrees to within a percent. Read independently for 08-20 it gives 118 attributed installs against our 122 registrations across all three Sinai arms, and $74.54 booked against our $74.13 — a 0.6% gap. Adjust's own figures for the day are identical at this cut and at the one this page first shipped on, so the spread closed from 4.8% entirely on our side, as the three unjoined payments above resolved. That is evidence the earlier gap was our missing joins and not a difference between two instruments. It is not in any table on this page.

4What would settle the open questions

Whether Vloom's sealed return recovers. One day at 0.56 against a series running 0.82–1.28 is an event; two is a level. Both frozen measures now agree the day fell and disagree only about how far — 46.5% sealed, 56.6% hour-neutral — so the question is no longer whether the fall was a maturity artefact. It is whether 08-21 comes back, and the same page for that day answers it.

Whether Sinai's first-day 2.06 survives its own payer count. It needs 16 payers before the number can be ranked against anything, and the 3.43 at the horizon rests on the same twelve people. At 12 in a day it reaches the bar within two days if delivery holds.

Whether the US-only accounts are worth their CPM. KBM1 and HR1 have now spent $128.94 on Vloom in a day at CPMs of $161.74 and $29.16. The valid comparison is against B1's own US cell, and the country block in Vloom's tab carries both sides of it.

Whether the instrument gate settles. 1.35 → 1.22 → 1.18 is a trend toward the band but is still outside it. Two candidates are unseparated — the FX table's age against the days it prices, and the fifth of API payments that join to no user — and the first is separable by stamping the table's own age against those days.

The three questions the 08-19 report handed forward all have answers. The click-to-install dip in hours 00, 07, 09, 12 and 23 did not repeat. Every one of those hours rose on 08-20, so that was a delivery window and not a decline. The instrument gate came in again, 1.22 to 1.18, which leaves it outside the 0.91–1.06 band for a third day while still moving the right way. And the new campaigns are no longer unknown ids: HR1's bailingxia_srfhjyd_ww_cvr_260819 is now enumerated as Vloom's, and the three _sn_ campaigns are enumerated as Sinai's, which is a different app on a different ledger.

vid_ww_bodysuit_10s_9x16_v1 still took no spend. B1's worldwide campaign has now run a single delivering ad for nine days.

5What this hands to the next read

The 08-21 report inherits a portfolio buying two apps in the same three accounts, which is a deliberate condition: Sinai exists to test whether a different client is a different product in the auction. This page cannot answer that on one day and 12 payers, but it establishes the measurement: both apps now run on the same instruments, on the same clock, with the same three measures, and neither one's money can enter the other's denominators.

It also hands forward a live question about B1. Vloom's delivery there fell away in the evening, impressions in hours 17 through 23 roughly halving against 08-19, on the same day Sinai began buying in that account. Whether those two facts are connected is exactly the competition question, and it needs the second day.

6Reproduce

This page is the second cut of 2026-08-20, taken once the day's cohort had matured. Run staged by python -m ad_ops.daily_pipeline --day 2026-08-20 into _runs/2026-08-22_2026-08-20_daily_auto, which pulled 16 Meta exports across five ad accounts at three grains, the Lum payment and registration feeds, and ran every instrument. Sinai's two feeds and its tables were added on top:

python ad_ops/lum_notifications/pull_payments_api.py --out <run>/sinai_backend \
    --app-source sin --start 2026-08-14 --end 2026-08-23
python ad_ops/lum_notifications/pull_registrations_sink.py --out <run>/sinai_lum \
    --app sin --start 2026-08-18T00:00:00Z --end 2026-08-22T04:00:00Z
python -m ad_ops.portfolio_split --run <run> --config run_config_sinai.json --app sinai_ww \
    --day 2026-08-20 --prev 2026-08-19

Payment cut 2026-08-22 00:46 UTC for Vloom and 03:03 UTC for Sinai; the first cut of this page ran at 2026-08-21 07:43 UTC, when the day's cohort was 34% covered on Vloom and 0% on Sinai's buy. Within-day and booked are unchanged between the two cuts except where a payment gained a payer to join to; the cohort columns are the ones that finished rising. Meta day 08-20 closed at 07:00 UTC on 08-21 on the UTC-7 accounts, and re-pulling a settled day moves spend by a few tenths of a percent — this cut has Vloom at $575.05 where the first had $574.13.

Report clock UTC throughout except the country section and the instrument gate, both labelled where they appear.

⚠⚠ THE ACCOUNTS DO NOT BUY THE SAME GEOGRAPHY, SO NO BLENDED FIGURE IS COMPARABLE ACROSS THEM. KBM1 and HR1 are targeted Inclusion: United States and deliver 100% US; B1 and the shared pair are worldwide and run 3-4% US. A US impression costs this portfolio roughly three times a blended one, so setting KBM1's CPM or cost per registration against B1's ACCOUNT figure reports a geography difference as an account difference. Compare a US-only account against B1's own US cell, which the country export carries. ⚠ Their campaign NAMES say ww; the names are wrong and the ad set is the authority.

accountgeographydayhourlycountryclicks / CTR / Meta installswhy
B1worldwide, ~4% USyesyesyesyesoffset k=7
Nomad Nodeworldwide, ~3% USyesyesyesyesoffset k=0
KBM1UNITED STATES ONLYyesyesyesyesoffset k=7
HR1UNITED STATES ONLYyesyesyesyesoffset k=0

Campaign ids in our records that belong to no known account. Each is counted in NO account, which is why the per-account rows can sum to less than the records hold. Add each to ACCOUNTS:

7Appendix — definitions

The report clock is UTC. Meta stamps its exports in each ad account's own zone: B1 and KBM1 are UTC-7, so a UTC day on either is stitched from two Meta days, hour by hour; HR1 and the two shared Nomad Node accounts are UTC+0, where the two days coincide. Stitching an account on the wrong offset moves its spend across the clock without changing any total, which is why an account whose offset is underived is held out of the hourly section.

The three revenue measures, never mixed and always labelled:

measurecountsmoves after the day closes?
bookedrevenue that arrived inside the day, whatever day its payer installed. Meta's own number is this basissettles about four hours after midnight, then fixed
within-dayinstalled that day and paid before that day closednever. Sealed at midnight
cohort at 24hthat day's installs, counted within 24 hours of each person's own installrises until every install has lived 24 hours, then frozen. Always quoted with coverage

All three are read on our own records: for Vloom the Lum ledger, for Sinai the in-app-purchase feed joined to the sink's registration push. Meta can express only booked, so it appears bracketed beside that column and never as a row of its own: quoting Meta's booked against our within-day would mix instruments and put part of the 0.91–1.06 instrument difference into a figure about the day.

Two install counts exist and are not interchangeable. Meta's Leads and our own registrations differ by 5–10%. Meta's is used for the delivery funnel, CPI and click-to-install, so numerator and denominator are both Meta's; ours is used for the payer rate and every revenue measure, so the payers and the installs they came from are the same population.

A regime is a stretch over which an ad set's daily budget did not change. It matters because the budget is a geography dial on this account: raising it buys wider and cheaper countries, so two creatives measured across a budget change are compared on different audiences.

The 16-payer bar. Below about 16 payers a revenue figure is a direction and never a magnitude. Cells under it are reported so the tables reconcile, and are not ranked.

Internal — noindex. Not for distribution outside the team.