Monthly campaign report — July 2026

Buying for this app began on 2026-07-23, so July is nine days here, 2026-07-23 to 2026-07-31, and nothing precedes it to compare against. B1 was the only ad account buying; the second began on 2026-08-11, after this period closes. Across the nine days B1 spent $2,455.00 over eight campaigns for 4,561 installs at $0.538 and 5,069 registrations, booked $3,125.48, sealed $2,344.02 inside the days that earned it, and its cohort at 24 hours is complete at $2,697.93. The account's own delivery export shows July spend this report does not count, because a test on another app and a third party's buy also ran on it; the amounts are in Reproduce. On the META clock our records read 1.07 of Meta's booked value, outside the documented 0.91 to 1.06 band. Definitions, the report clock and the three revenue measures are in the appendix.

1Overview

	2026-07-23 → 2026-07-31
days	9
spend	$2,455.00
impressions	347,058
CPM	$7.07
clicks	20,124
CTR	5.80%
installs (Meta)	4,561
click→install	22.7%
CPI	$0.538
registrations (ours)	5,069
booked	$3,125.48 (1.27) (Meta 1.18)
within-day	$2,344.02 (0.95)
cohort @ 24h	$2,697.93 (1.10) at 100%
within-day payers	233

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.

app	ARPU (d0)	ARPU (d3)	ARPU (d7)	ARPPU (d0)	ARPPU (d3)	ARPPU (d7)	installs
Vloom	$0.4861	$0.6372	$0.7713	$10.21	$11.35	$12.46	5,524

Campaign spend split across the nine days: T3 (Jul 25) 55.2%, the earner 14.6%, T2 (Jul 25) 10.9%, T1 (Jul 25) 8.3%, German (DE/AT/CH) 4.1%, T1 (Jul 23) 2.9%, T2 (Jul 23) 2.3%, T3 (Jul 23) 1.7%.

The month did not return its own money inside the day. Within-day is 0.95; the 1.27 booked is inherited cohort.

This is the only cohort reading in the series that is not a floor. H=24 is 100% covered and $2,697.93 is frozen.

One campaign is most of the month. T3 (Jul 25) took $1,355.94 of $2,455.00 and 284,604 of 347,058 impressions. The account rates are its own.

Cost per install spans two orders of magnitude across the eight buys, $0.324 to $18.098. A pooled account rate reads the test roster.

2Hourly, 2026-07-23 to 2026-07-31

The whole account on the report clock, pooled across the nine days. The hour axis is hour of day and not a timeline: each bucket holds nine hours of buying, one from each day, which is what makes the diurnal shape readable and what stops any single row being read as an event. Delivery columns are Meta's instrument on both sides of every ratio. Payer rate, ARPU and ARPPU are ours on both sides. The two install counts are never crossed. ours is our registrations in that bucket; cov is the bucket's cohort coverage at H=24.

hr	spend	impr	CPM	clicks	CTR	inst	clk→i	CPI	ours	cov	booked	bROAS	within	wROAS	wPay	cohort	cROAS
0	$52.96	6,100	8.68	426	6.98%	52	12.2%	$1.018	53	100%	$38.71	0.73	$0.00	0.00	0	$0.00	0.00
1	$74.99	11,489	6.53	640	5.57%	111	17.3%	$0.676	123	100%	$88.03	1.17	$52.96	0.71	9	$52.96	0.71
2	$85.60	12,061	7.10	644	5.34%	158	24.5%	$0.542	165	100%	$40.62	0.47	$68.16	0.80	8	$68.16	0.80
3	$78.74	12,005	6.56	629	5.24%	129	20.5%	$0.610	140	100%	$156.15	1.98	$166.43	2.11	15	$166.43	2.11
4	$85.43	15,385	5.55	814	5.29%	131	16.1%	$0.652	144	100%	$107.07	1.25	$90.90	1.06	13	$100.22	1.17
5	$109.12	18,425	5.92	1,003	5.44%	198	19.7%	$0.551	218	100%	$117.02	1.07	$131.53	1.21	16	$144.25	1.32
6	$116.83	21,582	5.41	1,079	5.00%	251	23.3%	$0.465	280	100%	$129.15	1.11	$132.99	1.14	13	$136.78	1.17
7	$106.53	18,695	5.70	937	5.01%	219	23.4%	$0.486	247	100%	$235.86	2.21	$183.85	1.73	21	$203.85	1.91
8	$116.36	21,051	5.53	1,049	4.98%	229	21.8%	$0.508	256	100%	$181.70	1.56	$170.89	1.47	16	$174.45	1.50
9	$115.87	19,698	5.88	1,050	5.33%	239	22.8%	$0.485	272	100%	$144.23	1.24	$41.77	0.36	5	$41.77	0.36
10	$119.25	18,397	6.48	1,042	5.66%	236	22.6%	$0.505	271	100%	$52.02	0.44	$39.47	0.33	5	$46.59	0.39
11	$116.37	16,121	7.22	885	5.49%	213	24.1%	$0.546	236	100%	$196.42	1.69	$225.41	1.94	15	$289.74	2.49
12	$107.50	13,499	7.96	792	5.87%	184	23.2%	$0.584	213	100%	$81.37	0.76	$94.78	0.88	10	$113.42	1.06
13	$111.16	14,218	7.82	826	5.81%	193	23.4%	$0.576	202	100%	$177.24	1.59	$63.93	0.58	9	$63.93	0.58
14	$129.34	14,017	9.23	838	5.98%	190	22.7%	$0.681	211	100%	$225.72	1.75	$211.67	1.64	12	$229.79	1.78
15	$141.13	17,713	7.97	1,116	6.30%	266	23.8%	$0.531	285	100%	$169.22	1.20	$146.26	1.04	13	$152.88	1.08
16	$165.90	23,992	6.91	1,373	5.72%	320	23.3%	$0.518	365	100%	$129.42	0.78	$110.24	0.66	7	$121.00	0.73
17	$149.58	21,681	6.90	1,338	6.17%	321	24.0%	$0.466	363	100%	$203.44	1.36	$143.32	0.96	16	$174.06	1.16
18	$132.89	18,533	7.17	1,146	6.18%	300	26.2%	$0.443	339	100%	$129.58	0.98	$80.00	0.60	10	$151.09	1.14
19	$93.47	12,340	7.57	866	7.02%	241	27.8%	$0.388	274	100%	$126.28	1.35	$61.93	0.66	7	$77.11	0.82
20	$83.63	6,532	12.80	516	7.90%	151	29.3%	$0.554	167	100%	$122.25	1.46	$59.46	0.71	4	$72.42	0.87
21	$52.16	5,008	10.42	386	7.71%	68	17.6%	$0.767	77	100%	$108.37	2.08	$25.22	0.48	3	$41.32	0.79
22	$47.69	3,492	13.66	322	9.22%	64	19.9%	$0.745	64	100%	$83.17	1.74	$32.09	0.67	4	$59.21	1.24
23	$62.50	5,024	12.44	407	8.10%	97	23.8%	$0.644	104	100%	$82.45	1.32	$10.75	0.17	2	$16.51	0.26
all	$2,455.00	347,058	7.07	20,124	5.80%	4,561	22.7%	$0.538	5,069	100%	$3,125.48	1.27	$2,344.02	0.95	233	$2,697.93	1.10

⚠ Within-day payers per bucket run 0 to 21, and only four buckets reach 16. Read the all row for level and the buckets for shape.

Coverage is 100% in every bucket: this month is read five weeks after it closed.

The two ends of the clock buy different things. CPM is 5.41 at hour 6 and 13.66 at hour 22; the click rate inverts that.

Installs are cheapest late afternoon and dearest around midnight, $0.388 at hour 19 against $1.018 at hour 0, while spend peaks at hour 16.

Four buckets clear the 16-payer bar and they do not agree: hours 5, 7, 8 and 17 return 1.21, 1.73, 1.47 and 0.96.

Against the period before it

There is no period before it. Buying began on 2026-07-23, which is this report's first day, so the matched-hours pairing, its sign tests and the chart that lays a prior period over this one have nothing to run against and none of them exists for this month. Everything in this section is a level read of nine days. The first comparison this series can make is August against July.

The period's motion: spend by hour, and impressions against installs

CTR, CPM, click-to-install and CPI by hour, whole account

3Country

Spend against within-day revenue, by country

Cost per install by country

(i) The portfolio by country, 2026-07-01..2026-07-31. 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.

cc	spend	share	impressions	CPM	registrations (ours)	cost/reg	within-day	wROAS	payers
IN	$1,474.61	60.1%	301,292	$4.89	3,597	$0.410	$1,561.56	1.06	215
MX	$236.12	9.6%	16,022	$14.74	295	$0.800	$162.29	0.69	21
US	$143.05	5.8%	1,575	$90.80	72	$1.987	$126.80	0.89	8
GB	$85.25	3.5%	1,362	$62.59	39	$2.186	$0.00	0.00	0
DE	$57.75	2.4%	1,049	$55.04	26	$2.221	$101.66	1.76	3
MY	$53.65	2.2%	3,419	$15.69	50	$1.073	$5.38	0.10	1
CH	$42.13	1.7%	1,011	$41.69	3	$14.044	$34.65	0.82	1
AU	$36.03	1.5%	578	$62.30	15	$2.402	$0.00	0.00	0
CA	$32.05	1.3%	669	$47.91	31	$1.034	$19.94	0.62	4
ID	$27.23	1.1%	3,947	$6.90	113	$0.241	$4.99	0.18	2
BR	$19.73	0.8%	1,369	$14.41	114	$0.173	$5.13	0.26	1
SA	$19.65	0.8%	1,073	$18.32	25	$0.786	$0.00	0.00	0

163 further countries are not listed, $227.74 between them (9.3% of this block).

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

(ii) The same cut per account, 2026-07-01..2026-07-31.

cc	spend	share	impressions	CPM	registrations (ours)	cost/reg	within-day	wROAS	payers
IN	$1,474.61	60.1%	301,292	$4.89	3,597	$0.410	$1,561.56	1.06	215
MX	$236.12	9.6%	16,022	$14.74	295	$0.800	$162.29	0.69	21
US	$143.05	5.8%	1,575	$90.80	72	$1.987	$126.80	0.89	8
GB	$85.25	3.5%	1,362	$62.59	39	$2.186	$0.00	0.00	0
DE	$57.75	2.4%	1,049	$55.04	26	$2.221	$101.66	1.76	3
MY	$53.65	2.2%	3,419	$15.69	50	$1.073	$5.38	0.10	1
CH	$42.13	1.7%	1,011	$41.69	3	$14.044	$34.65	0.82	1
AU	$36.03	1.5%	578	$62.30	15	$2.402	$0.00	0.00	0
CA	$32.05	1.3%	669	$47.91	31	$1.034	$19.94	0.62	4
ID	$27.23	1.1%	3,947	$6.90	113	$0.241	$4.99	0.18	2
BR	$19.73	0.8%	1,369	$14.41	114	$0.173	$5.13	0.26	1
SA	$19.65	0.8%	1,073	$18.32	25	$0.786	$0.00	0.00	0

163 further countries are not listed, $227.74 between them (9.3% 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.

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

Reconciliation over the period. Every per-country cell is an allocation across hours; every total is measured. Each account's raked country total equals its bought spend for these days: B1 $2,455.00. The per-day reconciliations, naming the Meta days each figure was raked from, are on the daily pages.

4Aggregate, the period

Pooled over the nine days on the UTC clock. Two things this section normally carries need a second period and there is none: the pooled matched-hours funnel with its two-proportion tests, and the mix-neutral account line, which holds one period's per-campaign figures to another's impression mix. Neither is printed. What remains is the level read of the whole month, which is the part that does not need a comparison to mean anything.

The three measures side by side

measure	2026-07-23 → 2026-07-31	moves after the period closes?
booked	$3,125.48 (1.27) (Meta 1.18)	settled
within-day	$2,344.02 (0.95)	never
cohort @ 24h	$2,697.93 (1.10) at 100%	frozen, fully covered

The three disagree by about a third of the month's revenue. Booked picks up earlier installs; cohort at 24 hours is hour-neutral; within-day cannot move.

Read as reference and never ranked, the same cohort to date is $4,862.31, a return of 1.98 at the 2026-08-16 payment cut. It says how much of this cohort's money arrived after its first day and it is not comparable to any other period.

Booked revenue by how long its payer had been installed

installed	2026-07-23 → 2026-07-31
the same day	$2,344.02 (75.0%)
one day earlier	$349.19 (11.2%)
two days earlier	$285.86 (9.1%)
three days earlier	$74.17 (2.4%)
four days earlier	$45.44 (1.5%)
five days earlier	$21.04 (0.7%)
six days earlier	$5.76 (0.2%)

The top row of this table is the within-day total, seen from the other side, which is why the gap between booked and within-day is exactly the inherited cohort and nothing else. Over a period the split stops being about which day money landed and becomes a decay curve.

Three quarters of what July booked came from same-day installs and the tail is gone inside three days — what H=24 was chosen against.

The three revenue measures on both periods, ROAS above each bar

Booked revenue by how long its payer had been installed

5The campaigns, side by side

Delivery is Meta's instrument on both sides of every ratio; every revenue measure is ours. Cohort horizon H=24h. Report clock UTC. Rank on CPM, cost per click and CPI only — geography is the treatment here, which is what makes those comparable across campaigns, while revenue still needs 16 payers (DECISION_LOG.md #37, #48) and a campaign under that bar carries a direction and never a magnitude.

campaign	spend	share	impressions	impr share	CPM	clicks	CTR	cost/click	installs	CPI	within-day	wROAS	payers
T3 (Jul 25, India/SEA)	$1,355.94	55.2%	284,604	82.0%	4.76	14,213	4.99%	$0.095	2,875	$0.472	$1,316.84	0.97	142
Earner (retired 08-03)	$358.20	14.6%	29,592	8.5%	12.10	3,423	11.57%	$0.105	1,106	$0.324	$591.31	1.65	61
T2 (Jul 25)	$268.63	10.9%	18,253	5.3%	14.72	1,092	5.98%	$0.246	298	$0.901	$144.06	0.54	16
T1 (Jul 25)	$203.93	8.3%	2,591	0.7%	78.71	354	13.66%	$0.576	102	$1.999	$87.87	0.43	6
German (DE/AT/CH)	$99.98	4.1%	2,140	0.6%	46.72	83	3.88%	$1.205	23	$4.347	$127.00	1.27	2
T1 (Jul 23)	$72.39	2.9%	1,296	0.4%	55.86	71	5.48%	$1.020	4	$18.098	$0.00	0.00	0
T2 (Jul 23)	$55.25	2.3%	2,680	0.8%	20.62	296	11.04%	$0.187	35	$1.579	$9.11	0.16	1
T3 (Jul 23)	$40.68	1.7%	5,902	1.7%	6.89	592	10.03%	$0.069	118	$0.345	$67.82	1.67	5

Every campaign in the table delivered. The instrument prints a booking-with-no-delivery line whenever a switched-off campaign lands conversions on spend it did not incur, and it printed none here, so the account totals and the eight tabs reconcile without an exclusion.

Price ran with the tier names and value ran against them — CPM 4.76 to 78.71, CPI $0.324 to $18.098, both ordered by tier.

Three campaigns reach 16 payers and only one earns: the earner at 1.65, against 0.97 and 0.54.

The earner is the only campaign that returned more than it cost, and it ran on two of the nine days.

T1 (Jul 23) bought 1,296 impressions for $72.39 and got four installs, at $18.098 each. It clears the tab bar on spend share alone.

The standing charts for this section plot spend share, CPM and CPI by campaign across the run. The chart generator emits none at this level, so section 5 carries its table alone.

6Inside the campaigns

This is the ad level. The account's whole asset roster comes first, ordered by spend across every campaign, and then one tab per campaign re-cuts the same battery inside it. The tabs are ordered by spend descending and every one of them carries the same four sections.

What a tab carries. Its charts are scoped to the campaign; its tables are asset-level only; and no claim lives only inside a tab.

A. Per asset — delivery

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

asset	spend	impressions	CPM	clicks	CTR	installs	click→install	CPI
`earner/bodysuit`	$312.93	26,159	11.96	2,968	11.35%	934	31.47%	$0.335
`T3b/utility_6s`	$305.33	48,353	6.31	2,653	5.49%	521	19.64%	$0.586
`T3b/sasian_spice5_15s`	$282.88	67,080	4.22	2,184	3.26%	453	20.74%	$0.624
`T3b/sasian_bodysuit_10s`	$281.69	54,387	5.18	2,836	5.21%	527	18.58%	$0.535
`T3b/sasian_bikini_6s`	$189.22	58,064	3.26	3,145	5.42%	754	23.97%	$0.251
`T3b/sasian_bikini_10s`	$120.14	34,303	3.50	1,841	5.37%	332	18.03%	$0.362
`T1b/utility_6s#2`	$74.18	877	84.58	148	16.88%	57	38.51%	$1.301
`T1b/utility_15s`	$73.98	888	83.31	127	14.30%	35	27.56%	$2.114
`T2b/utility_6s#3`	$68.17	3,568	19.11	244	6.84%	72	29.51%	$0.947
`T2b/utility_15s#2`	$52.84	3,199	16.52	170	5.31%	37	21.76%	$1.428
`T2b/latina_es_bikini_10s`	$42.73	4,706	9.08	291	6.18%	99	34.02%	$0.432
`T3b/sasian_slip_dress`	$39.93	5,257	7.60	406	7.72%	81	19.95%	$0.493
`earner/pool5`	$36.63	2,737	13.38	372	13.59%	144	38.71%	$0.254
`T1a/utility_sexy`	$30.97	559	55.40	24	4.29%	0	0.00%	—
`T1a/control_sexy`	$28.95	349	82.95	40	11.46%	4	10.00%	$7.237
`T3a/utility_sexy#2`	$28.90	4,173	6.93	478	11.45%	114	23.85%	$0.254
`T2a/control_sexy#2`	$28.18	1,303	21.63	121	9.29%	13	10.74%	$2.168
`DE/de_german_mature_6s`	$27.40	456	60.09	29	6.36%	8	27.59%	$3.425
`T2a/utility_sexy#3`	$27.07	1,377	19.66	175	12.71%	22	12.57%	$1.230
`T3b/sasian_fishnet_top_6s`	$25.15	4,135	6.08	167	4.04%	46	27.54%	$0.547
`DE/ch_arab_gulf_6s`	$21.75	498	43.67	12	2.41%	1	8.33%	$21.750
`T2b/latina_es_slip_dress`	$19.39	1,262	15.36	63	4.99%	27	42.86%	$0.718
`T2b/control_10s`	$14.50	910	15.93	55	6.04%	3	5.45%	$4.833
`T3b/sasian_bodysuit_6s`	$14.50	1,687	8.60	71	4.21%	15	21.13%	$0.967
`T1b/utility_10s`	$14.31	232	61.68	16	6.90%	0	0.00%	—
`T1b/control_6s`	$13.99	188	74.41	19	10.11%	1	5.26%	$13.990
`T1b/control_15s`	$13.83	213	64.93	19	8.92%	3	15.79%	$4.610
`T1b/control_10s#2`	$13.64	193	70.67	25	12.95%	6	24.00%	$2.273
`T2b/control_6s#2`	$13.35	817	16.34	41	5.02%	7	17.07%	$1.907
`T2b/control_15s#2`	$13.14	829	15.85	53	6.39%	6	11.32%	$2.190
`T3b/control_15s#3`	$13.01	1,483	8.77	124	8.36%	18	14.52%	$0.723
`T3b/utility_10s#2`	$12.79	1,287	9.94	138	10.72%	27	19.57%	$0.474
`T2b/utility_10s#3`	$12.79	628	20.37	44	7.01%	9	20.45%	$1.421
`T3b/utility_15s#3`	$12.66	1,137	11.13	99	8.71%	16	16.16%	$0.791
`T3b/control_10s#3`	$12.55	1,267	9.91	105	8.29%	10	9.52%	$1.255
`T3b/control_6s#3`	$12.18	1,579	7.71	164	10.39%	32	19.51%	$0.381
`T3a/control_sexy#3`	$11.78	1,729	6.81	114	6.59%	4	3.51%	$2.945
`T3b/sasian_corset_6s`	$11.62	2,059	5.64	132	6.41%	20	15.15%	$0.581
`T2b/latina_es_bodysuit_10s`	$11.01	822	13.39	55	6.69%	13	23.64%	$0.847
`T3b/sasian_offshoulder`	$9.32	1,060	8.79	71	6.70%	7	9.86%	$1.331
`earner/bikini`	$8.64	696	12.41	83	11.93%	28	33.73%	$0.309
`T3b/sasian_jean_shorts`	$6.68	828	8.07	46	5.56%	9	19.57%	$0.742
`T1a/control_sexy#4`	$6.56	178	36.85	4	2.25%	0	0.00%	—
`T2b/latina_es_offshoulder`	$6.04	308	19.61	11	3.57%	3	27.27%	$2.013
`T1a/utility_sexy#4`	$5.91	210	28.14	3	1.43%	0	0.00%	—
`DE/de_arab_gulf_10s`	$5.79	83	69.76	5	6.02%	1	20.00%	$5.790
`DE/ch_german_10s`	$5.20	102	50.98	2	1.96%	0	0.00%	—
`DE/ch_latina_true_6s`	$5.17	106	48.77	1	0.94%	0	0.00%	—
`DE/de_turkish_mature_10s`	$4.71	126	37.38	5	3.97%	3	60.00%	$1.570
`T2b/latina_es_bikini_6s`	$4.15	444	9.35	21	4.73%	7	33.33%	$0.593
`DE/german_6s`	$4.07	100	40.70	6	6.00%	0	0.00%	—
`DE/ch_german_mature_6s`	$4.02	103	39.03	0	0.00%	0	—	—
`DE/turkish_mature`	$3.95	57	69.30	4	7.02%	4	100.00%	$0.987
`DE/ch_arab_gulf_10s`	$3.66	146	25.07	1	0.68%	1	100.00%	$3.660
`T3b/sasian_fishnet_top_10s`	$3.63	350	10.37	12	3.43%	3	25.00%	$1.210
`T2b/latina_es_jean_shorts`	$3.26	234	13.93	13	5.56%	6	46.15%	$0.543
`DE/german_mature`	$3.01	89	33.82	6	6.74%	2	33.33%	$1.505
`T3b/sasian_corset_10s`	$2.66	288	9.24	19	6.60%	4	21.05%	$0.665
`DE/de_arab_gulf_6s`	$2.51	26	96.54	2	7.69%	1	50.00%	$2.510
`T2b/latina_es_corset_10s`	$2.25	192	11.72	12	6.25%	5	41.67%	$0.450
`T2b/latina_es_bodysuit_6s`	$2.00	137	14.60	10	7.30%	3	30.00%	$0.667
`DE/german`	$1.90	60	31.67	0	0.00%	0	—	—
`T2b/latina_es_spice5_15s`	$1.46	77	18.96	2	2.60%	1	50.00%	$1.460
`DE/latina_true`	$1.36	21	64.76	1	4.76%	0	0.00%	—
`DE/de_turkish_10s`	$1.16	30	38.67	6	20.00%	2	33.33%	$0.580
`DE/de_pool5_15s`	$1.04	21	49.52	0	0.00%	0	—	—
`DE/de_asian_mature_6s`	$0.99	37	26.76	0	0.00%	0	—	—
`T2b/latina_es_fishnet_top_6s`	$0.79	75	10.53	3	4.00%	0	0.00%	—
`DE/ch_german_mature_10s`	$0.76	30	25.33	1	3.33%	0	0.00%	—
`DE/turkish_6s`	$0.54	24	22.50	1	4.17%	0	0.00%	—
`T2b/latina_es_corset_6s`	$0.52	7	74.29	2	28.57%	0	0.00%	—
`DE/ch_latina_mature_10s`	$0.42	7	60.00	0	0.00%	0	—	—
`DE/de_latina_true_6s`	$0.25	5	50.00	0	0.00%	0	—	—
`T2b/latina_es_fishnet_top_10s`	$0.24	38	6.32	2	5.26%	0	0.00%	—
`DE/ch_german_6s`	$0.14	4	35.00	0	0.00%	0	—	—
`DE/ch_asian_mature_6s`	$0.11	3	36.67	1	33.33%	0	0.00%	—
`DE/ch_pool5_15s`	$0.07	4	17.50	0	0.00%	0	—	—

Six assets took most of the month and four of them are from the South Asian pack. Forty-six of the seventy-seven took under $13 each.

The two halves of the table are two different auctions. India creatives buy impressions at 3.26 to 5.18; T1a/ and T1b/ pay 70.67 to 84.58.

CTR by hour, one line per asset

Click-to-install by hour, one line per asset

Cost per install by hour, one line per asset

CPM by hour, one line per asset

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 period. Report clock UTC.

asset	booked	bROAS (Meta)	within-day	wROAS	cohort@H	cROAS	payers	payer%	ARPU	ARPPU	largest payer
`earner/bodysuit`	$508.49	1.62 (1.71)	$508.49	1.62	$603.14	1.93	50	5.0%	0.504	10.17	8.1%
`T3b/utility_6s`	$483.71	1.58 (1.11)	$226.49	0.74	$314.43	1.03	30	5.1%	0.383	7.55	10.4%
`T3b/sasian_spice5_15s`	$300.89	1.06 (0.96)	$217.76	0.77	$243.62	0.86	31	5.9%	0.416	7.02	5.9%
`T3b/sasian_bodysuit_10s`	$358.25	1.27 (1.35)	$280.05	0.99	$325.38	1.16	33	5.4%	0.461	8.49	14.8%
`T3b/sasian_bikini_6s`	$401.49	2.12 (1.87)	$283.40	1.50	$316.27	1.67	18	2.2%	0.341	15.74	33.1%
`T3b/sasian_bikini_10s`	$92.34	0.77 (0.99)	$79.46	0.66	$79.46	0.66	7	1.9%	0.215	11.35	52.0%
`T1b/utility_6s#2`	$66.93	0.90 (0.90)	$66.93	0.90	$66.93	0.90	3	5.2%	1.154	22.31	52.3%
`T1b/utility_15s`	$20.94	0.28 (0.28)	$20.94	0.28	$20.94	0.28	3	7.9%	0.551	6.98	52.4%
`T2b/utility_6s#3`	$27.34	0.40 (0.40)	$27.34	0.40	$27.34	0.40	3	4.0%	0.364	9.11	58.2%
`T2b/utility_15s#2`	$28.43	0.54 (0.64)	$21.63	0.41	$25.03	0.47	2	5.6%	0.601	10.81	73.6%
`T2b/latina_es_bikini_10s`	$90.02	2.11 (2.09)	$68.40	1.60	$71.80	1.68	9	9.2%	0.698	7.60	33.2%
`T3b/sasian_slip_dress`	$95.58	2.39 (1.83)	$67.63	1.69	$76.95	1.93	7	7.0%	0.676	9.66	31.0%
`earner/pool5`	$77.78	2.12 (2.18)	$77.78	2.12	$77.78	2.12	10	6.4%	0.499	7.78	23.1%
`T1a/utility_sexy`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—
`T1a/control_sexy`	$9.39	0.32 (0.32)	$0.00	0.00	$9.39	0.32	0	0.0%	0.000	—	—
`T3a/utility_sexy#2`	$93.57	3.24 (2.23)	$67.82	2.35	$71.38	2.47	5	3.7%	0.502	13.56	51.7%
`T2a/control_sexy#2`	$9.11	0.32 (0.32)	$9.11	0.32	$9.11	0.32	1	7.7%	0.701	9.11	100.0%
`DE/de_german_mature_6s`	$101.66	3.71 (3.36)	$92.36	3.37	$98.10	3.58	1	12.5%	11.545	92.36	100.0%
`T2a/utility_sexy#3`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`T3b/sasian_fishnet_top_6s`	$94.54	3.76 (5.02)	$88.78	3.53	$94.54	3.76	8	13.6%	1.505	11.10	46.6%
`DE/ch_arab_gulf_6s`	$47.02	2.16 (2.15)	$34.65	1.59	$47.02	2.16	1	50.0%	17.323	34.65	100.0%
`T2b/latina_es_slip_dress`	$5.71	0.29 (0.29)	$5.71	0.29	$5.71	0.29	1	3.1%	0.178	5.71	100.0%
`T2b/control_10s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`T3b/sasian_bodysuit_6s`	$43.13	2.97 (2.95)	$18.63	1.29	$27.95	1.93	2	11.8%	1.096	9.32	69.1%
`T1b/utility_10s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—
`T1b/control_6s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`T1b/control_15s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`T1b/control_10s#2`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`T2b/control_6s#2`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`T2b/control_15s#2`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`T3b/control_15s#3`	$9.32	0.72 (1.25)	$9.32	0.72	$9.32	0.72	1	4.8%	0.444	9.32	100.0%
`T3b/utility_10s#2`	$38.21	2.99 (0.89)	$11.52	0.90	$11.52	0.90	2	6.9%	0.397	5.76	50.0%
`T2b/utility_10s#3`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`T3b/utility_15s#3`	$10.75	0.85 (1.30)	$0.00	0.00	$4.99	0.39	0	0.0%	0.000	—	—
`T3b/control_10s#3`	$12.88	1.03 (0.00)	$3.56	0.28	$3.56	0.28	1	10.0%	0.356	3.56	100.0%
`T3b/control_6s#3`	$12.88	1.06 (0.00)	$5.76	0.47	$5.76	0.47	1	2.7%	0.156	5.76	100.0%
`T3a/control_sexy#3`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`T3b/sasian_corset_6s`	$24.50	2.11 (0.79)	$24.50	2.11	$24.50	2.11	1	4.5%	1.114	24.50	100.0%
`T2b/latina_es_bodysuit_10s`	$9.11	0.83 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`T3b/sasian_offshoulder`	$0.00	0.00 (2.12)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`earner/bikini`	$5.04	0.58 (0.58)	$5.04	0.58	$5.04	0.58	1	3.2%	0.163	5.04	100.0%
`T3b/sasian_jean_shorts`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`T1a/control_sexy#4`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—
`T2b/latina_es_offshoulder`	$46.48	7.70 (5.12)	$20.99	3.48	$20.99	3.48	1	33.3%	6.997	20.99	100.0%
`T1a/utility_sexy#4`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—
`DE/de_arab_gulf_10s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`DE/ch_german_10s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—
`DE/ch_latina_true_6s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—
`DE/de_turkish_mature_10s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`T2b/latina_es_bikini_6s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`DE/german_6s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—
`DE/ch_german_mature_6s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—
`DE/turkish_mature`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`DE/ch_arab_gulf_10s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`T3b/sasian_fishnet_top_10s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`T2b/latina_es_jean_shorts`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`DE/german_mature`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`T3b/sasian_corset_10s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`DE/de_arab_gulf_6s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`T2b/latina_es_corset_10s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`T2b/latina_es_bodysuit_6s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`DE/german`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—
`T2b/latina_es_spice5_15s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`DE/latina_true`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—
`DE/de_turkish_10s`	$0.00	0.00 (4.30)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`DE/de_pool5_15s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—
`DE/de_asian_mature_6s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—
`T2b/latina_es_fishnet_top_6s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—
`DE/ch_german_mature_10s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—
`DE/turkish_6s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—
`T2b/latina_es_corset_6s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—
`DE/ch_latina_mature_10s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—
`DE/de_latina_true_6s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—
`T2b/latina_es_fishnet_top_10s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—
`DE/ch_german_6s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—
`DE/ch_asian_mature_6s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—
`DE/ch_pool5_15s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—

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

Five assets in the whole month clear the bar, and four of them are in one campaign. Their within-day returns run 0.74 to 1.62.

earner/bodysuit is the least concentrated revenue row of the month and the only readable return above 1.00, its largest payer 8.1% against 33.1%.

The largest-payer column disqualifies most of the high returns on the page — 3.37, 3.48 and 3.53 each rest on one to eight payers.

Payer rate by day, one line per asset

ARPU by day, one line per asset

ARPPU by day, one line per asset

Within-day ROAS by day, one line per asset

Composite: every asset across every measure, normalised across the period

Money-chart markers are hollow where the point sits under 16 payers, and assets under 20 installs on the day are dropped from the composite and the region charts.

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.

asset	spend	India %	US %	other %	US % of its revenue
`earner/bodysuit`	$408.99	32.8%	4.7%	62.6%	8.6%
`T3b/utility_6s`	$305.33	98.9%	0.0%	1.1%	0.0%
`T3b/sasian_spice5_15s`	$282.88	100.0%	0.0%	0.0%	0.0%
`T3b/sasian_bodysuit_10s`	$281.69	100.0%	0.0%	0.0%	0.0%
`T3b/sasian_bikini_6s`	$189.22	100.0%	0.0%	0.0%	0.0%
`T3b/sasian_bikini_10s`	$120.14	100.0%	0.0%	0.0%	0.0%
`T1b/utility_6s#2`	$74.18	0.0%	48.4%	51.6%	100.0%
`T1b/utility_15s`	$73.98	0.0%	42.5%	57.5%	52.4%
`T2b/utility_6s#3`	$68.17	0.0%	0.0%	100.0%	0.0%
`T2b/utility_15s#2`	$52.84	0.0%	0.0%	100.0%	0.0%
`T2b/latina_es_bikini_10s`	$42.73	0.0%	0.0%	100.0%	0.0%
`T3b/sasian_slip_dress`	$39.93	100.0%	0.0%	0.0%	0.0%
`earner/pool5`	$36.63	27.3%	8.2%	64.5%	6.4%
`T1a/utility_sexy`	$30.97	0.0%	51.2%	48.8%	—
`T1a/control_sexy`	$28.95	0.0%	57.5%	42.5%	—
`T3a/utility_sexy#2`	$28.90	79.4%	0.0%	20.6%	0.0%
`T2a/control_sexy#2`	$28.18	0.0%	0.0%	100.0%	0.0%
`DE/de_german_mature_6s`	$27.40	0.0%	0.0%	100.0%	0.0%
`T2a/utility_sexy#3`	$27.07	0.0%	0.0%	100.0%	—
`T3b/sasian_fishnet_top_6s`	$25.15	100.0%	0.0%	0.0%	0.0%
`DE/ch_arab_gulf_6s`	$21.75	0.0%	0.0%	100.0%	0.0%
`T2b/latina_es_slip_dress`	$19.39	0.0%	0.0%	100.0%	0.0%
`T2b/control_10s`	$14.50	0.0%	0.0%	100.0%	—
`T3b/sasian_bodysuit_6s`	$14.50	100.0%	0.0%	0.0%	0.0%
`T1b/utility_10s`	$14.31	0.0%	46.3%	53.7%	—
`T1b/control_6s`	$13.99	0.0%	39.5%	60.5%	—
`T1b/control_15s`	$13.83	0.0%	37.4%	62.6%	—
`T1b/control_10s#2`	$13.64	0.0%	38.8%	61.2%	—
`T2b/control_6s#2`	$13.35	0.0%	0.0%	100.0%	—
`T2b/control_15s#2`	$13.14	0.0%	0.0%	100.0%	—
`T3b/control_15s#3`	$13.01	81.9%	0.0%	18.1%	0.0%
`T3b/utility_10s#2`	$12.79	82.5%	0.0%	17.5%	0.0%
`T2b/utility_10s#3`	$12.79	0.0%	0.0%	100.0%	—
`T3b/utility_15s#3`	$12.66	83.9%	0.0%	16.1%	—
`T3b/control_10s#3`	$12.55	76.6%	0.0%	23.4%	0.0%
`T3b/control_6s#3`	$12.18	61.7%	0.0%	38.3%	0.0%
`T3a/control_sexy#3`	$11.78	57.8%	0.0%	42.2%	—
`T3b/sasian_corset_6s`	$11.62	100.0%	0.0%	0.0%	0.0%
`T2b/latina_es_bodysuit_10s`	$11.01	0.0%	0.0%	100.0%	—
`T3b/sasian_offshoulder`	$9.32	100.0%	0.0%	0.0%	—
`earner/bikini`	$150.33	29.0%	5.0%	65.9%	0.0%
`T3b/sasian_jean_shorts`	$6.68	100.0%	0.0%	0.0%	—
`T1a/control_sexy#4`	$6.56	0.0%	21.2%	78.8%	—
`T2b/latina_es_offshoulder`	$6.04	0.0%	0.0%	100.0%	0.0%
`T1a/utility_sexy#4`	$5.91	0.0%	32.0%	68.0%	—
`DE/de_arab_gulf_10s`	$5.79	0.0%	0.0%	100.0%	—
`DE/ch_german_10s`	$5.20	0.0%	0.0%	100.0%	—
`DE/ch_latina_true_6s`	$5.17	0.0%	0.0%	100.0%	—
`DE/de_turkish_mature_10s`	$4.71	0.0%	0.0%	100.0%	—
`T2b/latina_es_bikini_6s`	$4.15	0.0%	0.0%	100.0%	—
`DE/german_6s`	$4.07	0.0%	0.0%	100.0%	—
`DE/ch_german_mature_6s`	$4.02	0.0%	0.0%	100.0%	—
`DE/turkish_mature`	$3.95	0.0%	0.0%	100.0%	—
`DE/ch_arab_gulf_10s`	$3.66	0.0%	0.0%	100.0%	—
`T3b/sasian_fishnet_top_10s`	$3.63	100.0%	0.0%	0.0%	—
`T2b/latina_es_jean_shorts`	$3.26	0.0%	0.0%	100.0%	—
`DE/german_mature`	$3.01	0.0%	0.0%	100.0%	—
`T3b/sasian_corset_10s`	$2.66	100.0%	0.0%	0.0%	—
`DE/de_arab_gulf_6s`	$2.51	0.0%	0.0%	100.0%	—
`T2b/latina_es_corset_10s`	$2.25	0.0%	0.0%	100.0%	—
`T2b/latina_es_bodysuit_6s`	$2.00	0.0%	0.0%	100.0%	—
`DE/german`	$1.90	0.0%	0.0%	100.0%	—
`T2b/latina_es_spice5_15s`	$1.46	0.0%	0.0%	100.0%	—
`DE/latina_true`	$1.36	0.0%	0.0%	100.0%	—
`DE/de_turkish_10s`	$1.16	0.0%	0.0%	100.0%	—
`DE/de_pool5_15s`	$1.04	0.0%	0.0%	100.0%	—
`DE/de_asian_mature_6s`	$0.99	0.0%	0.0%	100.0%	—
`T2b/latina_es_fishnet_top_6s`	$0.79	0.0%	0.0%	100.0%	—
`DE/ch_german_mature_10s`	$0.76	0.0%	0.0%	100.0%	—
`DE/turkish_6s`	$0.54	0.0%	0.0%	100.0%	—
`T2b/latina_es_corset_6s`	$0.52	0.0%	0.0%	100.0%	—
`DE/ch_latina_mature_10s`	$0.42	0.0%	0.0%	100.0%	—
`DE/de_latina_true_6s`	$0.25	0.0%	0.0%	100.0%	—
`T2b/latina_es_fishnet_top_10s`	$0.24	0.0%	0.0%	100.0%	—
`DE/ch_german_6s`	$0.14	0.0%	0.0%	100.0%	—
`DE/ch_asian_mature_6s`	$0.11	0.0%	0.0%	100.0%	—
`DE/ch_pool5_15s`	$0.07	0.0%	0.0%	100.0%	—

⚠ The seven-hour offset is not cosmetic on the earner rows. This table's window is the account's own days, which reach into 2026-08-01 UTC, where the earner's budget had been raised again. It shows earner/bodysuit at $408.99 and earner/bikini at $150.33 where table A shows $312.93 and $8.64. Read the percentages, never the dollars, and never against tables A and B.

T1b/utility_6s#2 bought a geography and earned there, on three payers — 48.4% of its money in the United States, 100.0% of its revenue from it.

Everything with real volume is an India buy — the T3b/sasian_ pack at 100.0% and T3b/utility_6s at 98.9%, none of it American.

Where each asset bought: share of its own spend by region

Cost per install by region, every asset

Within-day revenue by region, every asset

Showing

T3 (Jul 25, India/SEA) — every section below this line is T3 (Jul 25, India/SEA)'s: the day, the hours, the countries, the campaigns and the assets, counted on T3 (Jul 25, India/SEA)'s own ledger and its own accounts. Use the buttons above to switch.

The India and South-East Asia arm of the 2026-07-25 build, and the campaign that owns this month: $1,355.94 of the account's $2,455.00, 284,604 of its 347,058 impressions, 2,875 installs at $0.472 and 142 within-day payers at a return of 0.97. It is the only campaign whose payer count clears the 16-payer bar with room, and even inside it only four assets do.

Hourly, country and aggregate

The period's motion: spend by hour, and impressions against installs

CTR, CPM, click-to-install and CPI by hour, T3 (Jul 25, India/SEA)

Spend against within-day revenue, by country

Cost per install by country

The three revenue measures on both periods, ROAS above each bar

Booked revenue by how long its payer had been installed

A. Per asset — delivery

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

asset	spend	impressions	CPM	clicks	CTR	installs	click→install	CPI
`T3b/utility_6s`	$305.33	48,353	6.31	2,653	5.49%	521	19.64%	$0.586
`T3b/sasian_spice5_15s`	$282.88	67,080	4.22	2,184	3.26%	453	20.74%	$0.624
`T3b/sasian_bodysuit_10s`	$281.69	54,387	5.18	2,836	5.21%	527	18.58%	$0.535
`T3b/sasian_bikini_6s`	$189.22	58,064	3.26	3,145	5.42%	754	23.97%	$0.251
`T3b/sasian_bikini_10s`	$120.14	34,303	3.50	1,841	5.37%	332	18.03%	$0.362
`T3b/sasian_slip_dress`	$39.93	5,257	7.60	406	7.72%	81	19.95%	$0.493
`T3b/sasian_fishnet_top_6s`	$25.15	4,135	6.08	167	4.04%	46	27.54%	$0.547
`T3b/sasian_bodysuit_6s`	$14.50	1,687	8.60	71	4.21%	15	21.13%	$0.967
`T3b/control_15s#3`	$13.01	1,483	8.77	124	8.36%	18	14.52%	$0.723
`T3b/utility_10s#2`	$12.79	1,287	9.94	138	10.72%	27	19.57%	$0.474
`T3b/utility_15s#3`	$12.66	1,137	11.13	99	8.71%	16	16.16%	$0.791
`T3b/control_10s#3`	$12.55	1,267	9.91	105	8.29%	10	9.52%	$1.255
`T3b/control_6s#3`	$12.18	1,579	7.71	164	10.39%	32	19.51%	$0.381
`T3b/sasian_corset_6s`	$11.62	2,059	5.64	132	6.41%	20	15.15%	$0.581
`T3b/sasian_offshoulder`	$9.32	1,060	8.79	71	6.70%	7	9.86%	$1.331
`T3b/sasian_jean_shorts`	$6.68	828	8.07	46	5.56%	9	19.57%	$0.742
`T3b/sasian_fishnet_top_10s`	$3.63	350	10.37	12	3.43%	3	25.00%	$1.210
`T3b/sasian_corset_10s`	$2.66	288	9.24	19	6.60%	4	21.05%	$0.665

Five assets took nearly all of it and the campaign's own cost spread is wide: T3b/sasian_bikini_6s bought impressions at 3.26 and installs at $0.251 while T3b/sasian_offshoulder paid 8.79 and $1.331. The six utility_/control_ cells are the 07-25 build; everything named sasian_ is the twelve-creative South Asian pack added on 2026-07-26 into a pooled ad set, which is why the two groups are not a matched comparison.

CTR by hour, one line per asset

Click-to-install by hour, one line per asset

Cost per install by hour, one line per asset

CPM by hour, one line per asset

B. Per asset — the three revenue measures

asset	booked	bROAS (Meta)	within-day	wROAS	cohort@H	cROAS	payers	payer%	ARPU	ARPPU	largest payer
`T3b/utility_6s`	$483.71	1.58 (1.11)	$226.49	0.74	$314.43	1.03	30	5.1%	0.383	7.55	10.4%
`T3b/sasian_spice5_15s`	$300.89	1.06 (0.96)	$217.76	0.77	$243.62	0.86	31	5.9%	0.416	7.02	5.9%
`T3b/sasian_bodysuit_10s`	$358.25	1.27 (1.35)	$280.05	0.99	$325.38	1.16	33	5.4%	0.461	8.49	14.8%
`T3b/sasian_bikini_6s`	$401.49	2.12 (1.87)	$283.40	1.50	$316.27	1.67	18	2.2%	0.341	15.74	33.1%
`T3b/sasian_bikini_10s`	$92.34	0.77 (0.99)	$79.46	0.66	$79.46	0.66	7	1.9%	0.215	11.35	52.0%
`T3b/sasian_slip_dress`	$95.58	2.39 (1.83)	$67.63	1.69	$76.95	1.93	7	7.0%	0.676	9.66	31.0%
`T3b/sasian_fishnet_top_6s`	$94.54	3.76 (5.02)	$88.78	3.53	$94.54	3.76	8	13.6%	1.505	11.10	46.6%
`T3b/sasian_bodysuit_6s`	$43.13	2.97 (2.95)	$18.63	1.29	$27.95	1.93	2	11.8%	1.096	9.32	69.1%
`T3b/control_15s#3`	$9.32	0.72 (1.25)	$9.32	0.72	$9.32	0.72	1	4.8%	0.444	9.32	100.0%
`T3b/utility_10s#2`	$38.21	2.99 (0.89)	$11.52	0.90	$11.52	0.90	2	6.9%	0.397	5.76	50.0%
`T3b/utility_15s#3`	$10.75	0.85 (1.30)	$0.00	0.00	$4.99	0.39	0	0.0%	0.000	—	—
`T3b/control_10s#3`	$12.88	1.03 (0.00)	$3.56	0.28	$3.56	0.28	1	10.0%	0.356	3.56	100.0%
`T3b/control_6s#3`	$12.88	1.06 (0.00)	$5.76	0.47	$5.76	0.47	1	2.7%	0.156	5.76	100.0%
`T3b/sasian_corset_6s`	$24.50	2.11 (0.79)	$24.50	2.11	$24.50	2.11	1	4.5%	1.114	24.50	100.0%
`T3b/sasian_offshoulder`	$0.00	0.00 (2.12)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`T3b/sasian_jean_shorts`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`T3b/sasian_fishnet_top_10s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`T3b/sasian_corset_10s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—

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

Four rows clear the bar: T3b/sasian_bodysuit_10s at 33 payers and 0.99, T3b/sasian_spice5_15s at 31 and 0.77, T3b/utility_6s at 30 and 0.74, T3b/sasian_bikini_6s at 18 and 1.50. The highest of those is also the most concentrated, with 33.1% of its within-day revenue on one payer against 5.9% for T3b/sasian_spice5_15s, so the ordering between them is not settled by these numbers. T3b/sasian_fishnet_top_6s reads 3.53 on eight payers and is not readable at this sample size.

Payer rate by day, one line per asset

ARPU by day, one line per asset

ARPPU by day, one line per asset

Within-day ROAS by day, one line per asset

Composite: every asset across every measure, normalised across the period

C. Per asset — country mix

asset	spend	India %	US %	other %	US % of its revenue
`T3b/utility_6s`	$305.33	98.9%	0.0%	1.1%	0.0%
`T3b/sasian_spice5_15s`	$282.88	100.0%	0.0%	0.0%	0.0%
`T3b/sasian_bodysuit_10s`	$281.69	100.0%	0.0%	0.0%	0.0%
`T3b/sasian_bikini_6s`	$189.22	100.0%	0.0%	0.0%	0.0%
`T3b/sasian_bikini_10s`	$120.14	100.0%	0.0%	0.0%	0.0%
`T3b/sasian_slip_dress`	$39.93	100.0%	0.0%	0.0%	0.0%
`T3b/sasian_fishnet_top_6s`	$25.15	100.0%	0.0%	0.0%	0.0%
`T3b/sasian_bodysuit_6s`	$14.50	100.0%	0.0%	0.0%	0.0%
`T3b/control_15s#3`	$13.01	81.9%	0.0%	18.1%	0.0%
`T3b/utility_10s#2`	$12.79	82.5%	0.0%	17.5%	0.0%
`T3b/utility_15s#3`	$12.66	83.9%	0.0%	16.1%	—
`T3b/control_10s#3`	$12.55	76.6%	0.0%	23.4%	0.0%
`T3b/control_6s#3`	$12.18	61.7%	0.0%	38.3%	0.0%
`T3b/sasian_corset_6s`	$11.62	100.0%	0.0%	0.0%	0.0%
`T3b/sasian_offshoulder`	$9.32	100.0%	0.0%	0.0%	—
`T3b/sasian_jean_shorts`	$6.68	100.0%	0.0%	0.0%	—
`T3b/sasian_fishnet_top_10s`	$3.63	100.0%	0.0%	0.0%	—
`T3b/sasian_corset_10s`	$2.66	100.0%	0.0%	0.0%	—

The sasian_ pack is 100.0% India and the US column is 0.0% on every row, so nothing in table B here is a geography purchase. The five control_/utility_ cells run 61.7% to 98.9% India with the rest in the other T3 countries, which is the only mix difference the table can show and it is small.

Where each asset bought: share of its own spend by region

Cost per install by region, every asset

Within-day revenue by region, every asset

Earner (retired 08-03) — every section below this line is Earner (retired 08-03)'s: the day, the hours, the countries, the campaigns and the assets, counted on Earner (retired 08-03)'s own ledger and its own accounts. Use the buttons above to switch.

The worldwide CBO buy that opened on 2026-07-30 and was switched off on 2026-08-03. Inside this month it ran on two of the nine days for $358.20: 1,106 installs at $0.324, the cheapest in the period, and $591.31 within-day at 1.65 on 61 payers. DECISION_LOG.md #45 built it as an earner, one ad set with three creatives under CBO, which is why its per-asset rows are not a ranking.

Hourly, country and aggregate

The period's motion: spend by hour, and impressions against installs

CTR, CPM, click-to-install and CPI by hour, Earner (retired 08-03)

Spend against within-day revenue, by country

Cost per install by country

The three revenue measures on both periods, ROAS above each bar

Booked revenue by how long its payer had been installed

A. Per asset — delivery

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

asset	spend	impressions	CPM	clicks	CTR	installs	click→install	CPI
`earner/bodysuit`	$312.93	26,159	11.96	2,968	11.35%	934	31.47%	$0.335
`earner/pool5`	$36.63	2,737	13.38	372	13.59%	144	38.71%	$0.254
`earner/bikini`	$8.64	696	12.41	83	11.93%	28	33.73%	$0.309

CBO put $312.93 of the campaign's $358.20 on earner/bodysuit and $8.64 on earner/bikini. That is the concentration the format warns about at the ad set level: a set that spends this unevenly settled on one creative early and never ranked the three. The three are close on every delivery ratio, 11.96, 13.38 and 12.41 on CPM and $0.335, $0.254 and $0.309 on cost per install, so nothing in the allocation follows from what the table shows.

CTR by hour, one line per asset

Click-to-install by hour, one line per asset

Cost per install by hour, one line per asset

CPM by hour, one line per asset

B. Per asset — the three revenue measures

asset	booked	bROAS (Meta)	within-day	wROAS	cohort@H	cROAS	payers	payer%	ARPU	ARPPU	largest payer
`earner/bodysuit`	$508.49	1.62 (1.71)	$508.49	1.62	$603.14	1.93	50	5.0%	0.504	10.17	8.1%
`earner/pool5`	$77.78	2.12 (2.18)	$77.78	2.12	$77.78	2.12	10	6.4%	0.499	7.78	23.1%
`earner/bikini`	$5.04	0.58 (0.58)	$5.04	0.58	$5.04	0.58	1	3.2%	0.163	5.04	100.0%

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

Only earner/bodysuit clears the bar, at 50 payers, and its largest single payer is 8.1% of its own within-day revenue. That combination makes its 1.62 the one per-asset return in the month worth treating as a level. earner/pool5 reads 2.12 on ten payers and earner/bikini 0.58 on one; neither is readable at this sample size. Budget moved under all three inside the two days they ran, from $150 a day to $250, $500 and $875, so none of this is regime-matched either.

Payer rate by day, one line per asset

ARPU by day, one line per asset

ARPPU by day, one line per asset

Within-day ROAS by day, one line per asset

Composite: every asset across every measure, normalised across the period

C. Per asset — country mix

asset	spend	India %	US %	other %	US % of its revenue
`earner/bodysuit`	$408.99	32.8%	4.7%	62.6%	8.6%
`earner/pool5`	$36.63	27.3%	8.2%	64.5%	6.4%
`earner/bikini`	$150.33	29.0%	5.0%	65.9%	0.0%

⚠ This is where the clock offset does real damage. The account's own days reach seven hours past the report's, into 2026-08-01 UTC, where this campaign's budget had been raised again. The spend column here reads $408.99 and $150.33 against table A's $312.93 and $8.64, so the dollars are a different window's and only the percentages carry across.

The mix itself is the point of an unrestricted worldwide buy: about a third India, a twentieth United States and the rest spread, on all three creatives. The US share of earner/bodysuit's own revenue is 8.6% against 4.7% of its spend, which is the mildest version of that gap on the page.

Where each asset bought: share of its own spend by region

Cost per install by region, every asset

Within-day revenue by region, every asset

T2 (Jul 25) — every section below this line is T2 (Jul 25)'s: the day, the hours, the countries, the campaigns and the assets, counted on T2 (Jul 25)'s own ledger and its own accounts. Use the buttons above to switch.

The mid-tier arm of the 07-25 build, buying the United Arab Emirates, Saudi Arabia, Malaysia, South Africa and Mexico, plus the Spanish-language Latina pack added to it on 2026-07-26. $268.63 for 18,253 impressions, 298 installs at $0.901 and $144.06 within-day at 0.54, on exactly 16 payers, which puts the campaign exactly on the bar.

Hourly, country and aggregate

The period's motion: spend by hour, and impressions against installs

CTR, CPM, click-to-install and CPI by hour, T2 (Jul 25)

Spend against within-day revenue, by country

Cost per install by country

The three revenue measures on both periods, ROAS above each bar

Booked revenue by how long its payer had been installed

A. Per asset — delivery

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

asset	spend	impressions	CPM	clicks	CTR	installs	click→install	CPI
`T2b/utility_6s#3`	$68.17	3,568	19.11	244	6.84%	72	29.51%	$0.947
`T2b/utility_15s#2`	$52.84	3,199	16.52	170	5.31%	37	21.76%	$1.428
`T2b/latina_es_bikini_10s`	$42.73	4,706	9.08	291	6.18%	99	34.02%	$0.432
`T2b/latina_es_slip_dress`	$19.39	1,262	15.36	63	4.99%	27	42.86%	$0.718
`T2b/control_10s`	$14.50	910	15.93	55	6.04%	3	5.45%	$4.833
`T2b/control_6s#2`	$13.35	817	16.34	41	5.02%	7	17.07%	$1.907
`T2b/control_15s#2`	$13.14	829	15.85	53	6.39%	6	11.32%	$2.190
`T2b/utility_10s#3`	$12.79	628	20.37	44	7.01%	9	20.45%	$1.421
`T2b/latina_es_bodysuit_10s`	$11.01	822	13.39	55	6.69%	13	23.64%	$0.847
`T2b/latina_es_offshoulder`	$6.04	308	19.61	11	3.57%	3	27.27%	$2.013
`T2b/latina_es_bikini_6s`	$4.15	444	9.35	21	4.73%	7	33.33%	$0.593
`T2b/latina_es_jean_shorts`	$3.26	234	13.93	13	5.56%	6	46.15%	$0.543
`T2b/latina_es_corset_10s`	$2.25	192	11.72	12	6.25%	5	41.67%	$0.450
`T2b/latina_es_bodysuit_6s`	$2.00	137	14.60	10	7.30%	3	30.00%	$0.667
`T2b/latina_es_spice5_15s`	$1.46	77	18.96	2	2.60%	1	50.00%	$1.460
`T2b/latina_es_fishnet_top_6s`	$0.79	75	10.53	3	4.00%	0	0.00%	—
`T2b/latina_es_corset_6s`	$0.52	7	74.29	2	28.57%	0	0.00%	—
`T2b/latina_es_fishnet_top_10s`	$0.24	38	6.32	2	5.26%	0	0.00%	—

The Latina pack buys better than the tier build it was added to — T2b/latina_es_bikini_10s at 9.08 and $0.432 against T2b/utility_6s#3 at 19.11 and $0.947.

CTR by hour, one line per asset

Click-to-install by hour, one line per asset

Cost per install by hour, one line per asset

CPM by hour, one line per asset

B. Per asset — the three revenue measures

asset	booked	bROAS (Meta)	within-day	wROAS	cohort@H	cROAS	payers	payer%	ARPU	ARPPU	largest payer
`T2b/utility_6s#3`	$27.34	0.40 (0.40)	$27.34	0.40	$27.34	0.40	3	4.0%	0.364	9.11	58.2%
`T2b/utility_15s#2`	$28.43	0.54 (0.64)	$21.63	0.41	$25.03	0.47	2	5.6%	0.601	10.81	73.6%
`T2b/latina_es_bikini_10s`	$90.02	2.11 (2.09)	$68.40	1.60	$71.80	1.68	9	9.2%	0.698	7.60	33.2%
`T2b/latina_es_slip_dress`	$5.71	0.29 (0.29)	$5.71	0.29	$5.71	0.29	1	3.1%	0.178	5.71	100.0%
`T2b/control_10s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`T2b/control_6s#2`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`T2b/control_15s#2`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`T2b/utility_10s#3`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`T2b/latina_es_bodysuit_10s`	$9.11	0.83 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`T2b/latina_es_offshoulder`	$46.48	7.70 (5.12)	$20.99	3.48	$20.99	3.48	1	33.3%	6.997	20.99	100.0%
`T2b/latina_es_bikini_6s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`T2b/latina_es_jean_shorts`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`T2b/latina_es_corset_10s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`T2b/latina_es_bodysuit_6s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`T2b/latina_es_spice5_15s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`T2b/latina_es_fishnet_top_6s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—
`T2b/latina_es_corset_6s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—
`T2b/latina_es_fishnet_top_10s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—

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

No asset here clears the bar. The campaign's 16 payers are spread across five creatives, the largest holding nine, and twelve of the eighteen assets have no revenue on any measure. The two figures that look like results, 3.48 on T2b/latina_es_offshoulder and 1.60 on T2b/latina_es_bikini_10s, rest on one payer and on nine with a third of the money on one of them. Not readable at this sample size.

Payer rate by day, one line per asset

ARPU by day, one line per asset

ARPPU by day, one line per asset

Within-day ROAS by day, one line per asset

Composite: every asset across every measure, normalised across the period

C. Per asset — country mix

asset	spend	India %	US %	other %	US % of its revenue
`T2b/utility_6s#3`	$68.17	0.0%	0.0%	100.0%	0.0%
`T2b/utility_15s#2`	$52.84	0.0%	0.0%	100.0%	0.0%
`T2b/latina_es_bikini_10s`	$42.73	0.0%	0.0%	100.0%	0.0%
`T2b/latina_es_slip_dress`	$19.39	0.0%	0.0%	100.0%	0.0%
`T2b/control_10s`	$14.50	0.0%	0.0%	100.0%	—
`T2b/control_6s#2`	$13.35	0.0%	0.0%	100.0%	—
`T2b/control_15s#2`	$13.14	0.0%	0.0%	100.0%	—
`T2b/utility_10s#3`	$12.79	0.0%	0.0%	100.0%	—
`T2b/latina_es_bodysuit_10s`	$11.01	0.0%	0.0%	100.0%	—
`T2b/latina_es_offshoulder`	$6.04	0.0%	0.0%	100.0%	0.0%
`T2b/latina_es_bikini_6s`	$4.15	0.0%	0.0%	100.0%	—
`T2b/latina_es_jean_shorts`	$3.26	0.0%	0.0%	100.0%	—
`T2b/latina_es_corset_10s`	$2.25	0.0%	0.0%	100.0%	—
`T2b/latina_es_bodysuit_6s`	$2.00	0.0%	0.0%	100.0%	—
`T2b/latina_es_spice5_15s`	$1.46	0.0%	0.0%	100.0%	—
`T2b/latina_es_fishnet_top_6s`	$0.79	0.0%	0.0%	100.0%	—
`T2b/latina_es_corset_6s`	$0.52	0.0%	0.0%	100.0%	—
`T2b/latina_es_fishnet_top_10s`	$0.24	0.0%	0.0%	100.0%	—

Every row is 0.0% India, 0.0% United States and 100.0% other, which is this campaign's geo definition working as set. The column cannot separate Mexico from Saudi Arabia, so the mix question here is answered in section 3, where Mexico took $243.10 at 0.67 and Malaysia $59.46 at 0.09.

Where each asset bought: share of its own spend by region

Cost per install by region, every asset

Within-day revenue by region, every asset

T1 (Jul 25) — every section below this line is T1 (Jul 25)'s: the day, the hours, the countries, the campaigns and the assets, counted on T1 (Jul 25)'s own ledger and its own accounts. Use the buttons above to switch.

The English-speaking arm of the 07-25 build: the United States, Canada, the United Kingdom, Australia and New Zealand. $203.93 bought 2,591 impressions at a CPM of 78.71 and 102 installs at $1.999, with $87.87 within-day at 0.43 on six payers.

Hourly, country and aggregate

The period's motion: spend by hour, and impressions against installs

CTR, CPM, click-to-install and CPI by hour, T1 (Jul 25)

Spend against within-day revenue, by country

Cost per install by country

The three revenue measures on both periods, ROAS above each bar

Booked revenue by how long its payer had been installed

A. Per asset — delivery

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

asset	spend	impressions	CPM	clicks	CTR	installs	click→install	CPI
`T1b/utility_6s#2`	$74.18	877	84.58	148	16.88%	57	38.51%	$1.301
`T1b/utility_15s`	$73.98	888	83.31	127	14.30%	35	27.56%	$2.114
`T1b/utility_10s`	$14.31	232	61.68	16	6.90%	0	0.00%	—
`T1b/control_6s`	$13.99	188	74.41	19	10.11%	1	5.26%	$13.990
`T1b/control_15s`	$13.83	213	64.93	19	8.92%	3	15.79%	$4.610
`T1b/control_10s#2`	$13.64	193	70.67	25	12.95%	6	24.00%	$2.273

The click rates are the best in the month and they buy nothing: T1b/utility_6s#2 clicked 16.88% and paid $1.301 an install at 84.58 CPM.

CTR by hour, one line per asset

Click-to-install by hour, one line per asset

Cost per install by hour, one line per asset

CPM by hour, one line per asset

B. Per asset — the three revenue measures

asset	booked	bROAS (Meta)	within-day	wROAS	cohort@H	cROAS	payers	payer%	ARPU	ARPPU	largest payer
`T1b/utility_6s#2`	$66.93	0.90 (0.90)	$66.93	0.90	$66.93	0.90	3	5.2%	1.154	22.31	52.3%
`T1b/utility_15s`	$20.94	0.28 (0.28)	$20.94	0.28	$20.94	0.28	3	7.9%	0.551	6.98	52.4%
`T1b/utility_10s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—
`T1b/control_6s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`T1b/control_15s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`T1b/control_10s#2`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—

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

Six payers in the campaign, three on each utility_ cell and none anywhere else. Their 0.90 and 0.28 each rest on a single payer holding 52.3% and 52.4% of the cell's revenue, so the ordering between the two is one person. Not readable at this sample size.

Payer rate by day, one line per asset

ARPU by day, one line per asset

ARPPU by day, one line per asset

Within-day ROAS by day, one line per asset

Composite: every asset across every measure, normalised across the period

C. Per asset — country mix

asset	spend	India %	US %	other %	US % of its revenue
`T1b/utility_6s#2`	$74.18	0.0%	48.4%	51.6%	100.0%
`T1b/utility_15s`	$73.98	0.0%	42.5%	57.5%	52.4%
`T1b/utility_10s`	$14.31	0.0%	46.3%	53.7%	—
`T1b/control_6s`	$13.99	0.0%	39.5%	60.5%	—
`T1b/control_15s`	$13.83	0.0%	37.4%	62.6%	—
`T1b/control_10s#2`	$13.64	0.0%	38.8%	61.2%	—

T1b/utility_6s#2 spent 48.4% of its money in the United States and took 100.0% of its within-day revenue from there. That is three payers landing in one country while the other four had none, which is too thin to read as a fact about the American market.

Where each asset bought: share of its own spend by region

Cost per install by region, every asset

Within-day revenue by region, every asset

German (DE/AT/CH) — every section below this line is German (DE/AT/CH)'s: the day, the hours, the countries, the campaigns and the assets, counted on German (DE/AT/CH)'s own ledger and its own accounts. Use the buttons above to switch.

The DACH campaign, published and switched on 2026-07-30 with a DE+AT ad set and a CH ad set at $75 a day each. Inside this month it ran on two of the nine days for $99.98: 2,140 impressions at a CPM of 46.72, 23 installs at $4.347 and $127.00 within-day at 1.27, on two payers. It is the one campaign in the period whose return is above 1.00 and unreadable at the same time.

Hourly, country and aggregate

The period's motion: spend by hour, and impressions against installs

CTR, CPM, click-to-install and CPI by hour, German (DE/AT/CH)

Spend against within-day revenue, by country

Cost per install by country

The three revenue measures on both periods, ROAS above each bar

Booked revenue by how long its payer had been installed

A. Per asset — delivery

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

asset	spend	impressions	CPM	clicks	CTR	installs	click→install	CPI
`DE/de_german_mature_6s`	$27.40	456	60.09	29	6.36%	8	27.59%	$3.425
`DE/ch_arab_gulf_6s`	$21.75	498	43.67	12	2.41%	1	8.33%	$21.750
`DE/de_arab_gulf_10s`	$5.79	83	69.76	5	6.02%	1	20.00%	$5.790
`DE/ch_german_10s`	$5.20	102	50.98	2	1.96%	0	0.00%	—
`DE/ch_latina_true_6s`	$5.17	106	48.77	1	0.94%	0	0.00%	—
`DE/de_turkish_mature_10s`	$4.71	126	37.38	5	3.97%	3	60.00%	$1.570
`DE/german_6s`	$4.07	100	40.70	6	6.00%	0	0.00%	—
`DE/ch_german_mature_6s`	$4.02	103	39.03	0	0.00%	0	—	—
`DE/turkish_mature`	$3.95	57	69.30	4	7.02%	4	100.00%	$0.987
`DE/ch_arab_gulf_10s`	$3.66	146	25.07	1	0.68%	1	100.00%	$3.660
`DE/german_mature`	$3.01	89	33.82	6	6.74%	2	33.33%	$1.505
`DE/de_arab_gulf_6s`	$2.51	26	96.54	2	7.69%	1	50.00%	$2.510
`DE/german`	$1.90	60	31.67	0	0.00%	0	—	—
`DE/latina_true`	$1.36	21	64.76	1	4.76%	0	0.00%	—
`DE/de_turkish_10s`	$1.16	30	38.67	6	20.00%	2	33.33%	$0.580
`DE/de_pool5_15s`	$1.04	21	49.52	0	0.00%	0	—	—
`DE/de_asian_mature_6s`	$0.99	37	26.76	0	0.00%	0	—	—
`DE/ch_german_mature_10s`	$0.76	30	25.33	1	3.33%	0	0.00%	—
`DE/turkish_6s`	$0.54	24	22.50	1	4.17%	0	0.00%	—
`DE/ch_latina_mature_10s`	$0.42	7	60.00	0	0.00%	0	—	—
`DE/de_latina_true_6s`	$0.25	5	50.00	0	0.00%	0	—	—
`DE/ch_german_6s`	$0.14	4	35.00	0	0.00%	0	—	—
`DE/ch_asian_mature_6s`	$0.11	3	36.67	1	33.33%	0	0.00%	—
`DE/ch_pool5_15s`	$0.07	4	17.50	0	0.00%	0	—	—

Twenty-four creatives shared $99.98 and nineteen got under $5. Only the two largest have a readable click rate, 6.36% against 2.41%.

CTR by hour, one line per asset

Click-to-install by hour, one line per asset

Cost per install by hour, one line per asset

CPM by hour, one line per asset

B. Per asset — the three revenue measures

asset	booked	bROAS (Meta)	within-day	wROAS	cohort@H	cROAS	payers	payer%	ARPU	ARPPU	largest payer
`DE/de_german_mature_6s`	$101.66	3.71 (3.36)	$92.36	3.37	$98.10	3.58	1	12.5%	11.545	92.36	100.0%
`DE/ch_arab_gulf_6s`	$47.02	2.16 (2.15)	$34.65	1.59	$47.02	2.16	1	50.0%	17.323	34.65	100.0%
`DE/de_arab_gulf_10s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`DE/ch_german_10s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—
`DE/ch_latina_true_6s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—
`DE/de_turkish_mature_10s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`DE/german_6s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—
`DE/ch_german_mature_6s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—
`DE/turkish_mature`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`DE/ch_arab_gulf_10s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`DE/german_mature`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`DE/de_arab_gulf_6s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`DE/german`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—
`DE/latina_true`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—
`DE/de_turkish_10s`	$0.00	0.00 (4.30)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—
`DE/de_pool5_15s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—
`DE/de_asian_mature_6s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—
`DE/ch_german_mature_10s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—
`DE/turkish_6s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—
`DE/ch_latina_mature_10s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—
`DE/de_latina_true_6s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—
`DE/ch_german_6s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—
`DE/ch_asian_mature_6s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—
`DE/ch_pool5_15s`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	—	—	—	—

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

Two payers in the whole campaign, one on each of two creatives, and each is 100.0% of that creative's revenue. The 3.37 and the 1.59 are two people, and the campaign's 1.27 in section 5 is the same two people one level up. Not readable at this sample size. What the July DACH read did establish sits in DECISION_LOG.md #44 and rests on delivery: German-area impressions cost several times India's and the geography earns enough per install to survive that, while no creative in the pack reached break-even.

Payer rate by day, one line per asset

ARPU by day, one line per asset

ARPPU by day, one line per asset

Within-day ROAS by day, one line per asset

Composite: every asset across every measure, normalised across the period

C. Per asset — country mix

asset	spend	India %	US %	other %	US % of its revenue
`DE/de_german_mature_6s`	$27.40	0.0%	0.0%	100.0%	0.0%
`DE/ch_arab_gulf_6s`	$21.75	0.0%	0.0%	100.0%	0.0%
`DE/de_arab_gulf_10s`	$5.79	0.0%	0.0%	100.0%	—
`DE/ch_german_10s`	$5.20	0.0%	0.0%	100.0%	—
`DE/ch_latina_true_6s`	$5.17	0.0%	0.0%	100.0%	—
`DE/de_turkish_mature_10s`	$4.71	0.0%	0.0%	100.0%	—
`DE/german_6s`	$4.07	0.0%	0.0%	100.0%	—
`DE/ch_german_mature_6s`	$4.02	0.0%	0.0%	100.0%	—
`DE/turkish_mature`	$3.95	0.0%	0.0%	100.0%	—
`DE/ch_arab_gulf_10s`	$3.66	0.0%	0.0%	100.0%	—
`DE/german_mature`	$3.01	0.0%	0.0%	100.0%	—
`DE/de_arab_gulf_6s`	$2.51	0.0%	0.0%	100.0%	—
`DE/german`	$1.90	0.0%	0.0%	100.0%	—
`DE/latina_true`	$1.36	0.0%	0.0%	100.0%	—
`DE/de_turkish_10s`	$1.16	0.0%	0.0%	100.0%	—
`DE/de_pool5_15s`	$1.04	0.0%	0.0%	100.0%	—
`DE/de_asian_mature_6s`	$0.99	0.0%	0.0%	100.0%	—
`DE/ch_german_mature_10s`	$0.76	0.0%	0.0%	100.0%	—
`DE/turkish_6s`	$0.54	0.0%	0.0%	100.0%	—
`DE/ch_latina_mature_10s`	$0.42	0.0%	0.0%	100.0%	—
`DE/de_latina_true_6s`	$0.25	0.0%	0.0%	100.0%	—
`DE/ch_german_6s`	$0.14	0.0%	0.0%	100.0%	—
`DE/ch_asian_mature_6s`	$0.11	0.0%	0.0%	100.0%	—
`DE/ch_pool5_15s`	$0.07	0.0%	0.0%	100.0%	—

Every row is 100.0% other, as the campaign's targeting requires. Section 3 carries the split this table cannot: Germany $60.82 at 1.67 and Switzerland $43.05 at 0.80, on one payer each.

Where each asset bought: share of its own spend by region

Cost per install by region, every asset

Within-day revenue by region, every asset

T1 (Jul 23) — every section below this line is T1 (Jul 23)'s: the day, the hours, the countries, the campaigns and the assets, counted on T1 (Jul 23)'s own ledger and its own accounts. Use the buttons above to switch.

The first-round English-speaking arm, from the 2026-07-23 build. It took $72.39 and 1,296 impressions at a CPM of 55.86 and produced four installs, at $18.098 each. It has no payers and $0.00 on all three revenue measures. It clears the tab bar on spend share and on nothing else: this is a campaign that bought impressions and almost nothing after them.

Hourly, country and aggregate

The period's motion: spend by hour, and impressions against installs

CTR, CPM, click-to-install and CPI by hour, T1 (Jul 23)

Spend against within-day revenue, by country

Cost per install by country

The three revenue measures on both periods, ROAS above each bar

Booked revenue by how long its payer had been installed

A. Per asset — delivery

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

asset	spend	impressions	CPM	clicks	CTR	installs	click→install	CPI
`T1a/utility_sexy`	$30.97	559	55.40	24	4.29%	0	0.00%	—
`T1a/control_sexy`	$28.95	349	82.95	40	11.46%	4	10.00%	$7.237
`T1a/control_sexy#4`	$6.56	178	36.85	4	2.25%	0	0.00%	—
`T1a/utility_sexy#4`	$5.91	210	28.14	3	1.43%	0	0.00%	—

Three of the four creatives produced no installs at all. The fourth, T1a/control_sexy, produced four at $7.237, from 349 impressions bought at 82.95 per thousand. The two arms that were meant to be compared, utility_sexy and control_sexy, took $30.97 and $28.95 and returned zero installs against four, a difference of four events.

CTR by hour, one line per asset

Click-to-install by hour, one line per asset

Cost per install by hour, one line per asset

CPM by hour, one line per asset

B. Per asset — the three revenue measures

asset	booked	bROAS (Meta)	within-day	cohort@H	cROAS	payer%	ARPU	ARPPU	largest payer
`T1a/utility_sexy`	$0.00	0.00 (0.00)	$0.00	$0.00	0.00	—	—	—	—
`T1a/control_sexy`	$9.39	0.32 (0.32)	$0.00	$9.39	0.32	0.0%	0.000	—	—
`T1a/control_sexy#4`	$0.00	0.00 (0.00)	$0.00	$0.00	0.00	—	—	—	—
`T1a/utility_sexy#4`	$0.00	0.00 (0.00)	$0.00	$0.00	0.00	—	—	—	—

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

$9.39 booked against $72.39 of spend, none of it within-day, and no payers anywhere in the campaign. The usual caution does not apply here, because there is no figure to be cautious about.

Payer rate by day, one line per asset

ARPU by day, one line per asset

ARPPU by day, one line per asset

Within-day ROAS by day, one line per asset

Composite: every asset across every measure, normalised across the period

C. Per asset — country mix

asset	spend	India %	US %	other %	US % of its revenue
`T1a/utility_sexy`	$30.97	0.0%	51.2%	48.8%	—
`T1a/control_sexy`	$28.95	0.0%	57.5%	42.5%	—
`T1a/control_sexy#4`	$6.56	0.0%	21.2%	78.8%	—
`T1a/utility_sexy#4`	$5.91	0.0%	32.0%	68.0%	—

Between a fifth and three fifths of each creative's money went to the United States and the last column is empty on every row, because there was no revenue to attribute anywhere.

Where each asset bought: share of its own spend by region

Cost per install by region, every asset

Within-day revenue by region, every asset

T2 (Jul 23) — every section below this line is T2 (Jul 23)'s: the day, the hours, the countries, the campaigns and the assets, counted on T2 (Jul 23)'s own ledger and its own accounts. Use the buttons above to switch.

The first-round mid-tier arm. $55.25 for 2,680 impressions at a CPM of 20.62, 35 installs at $1.579 and $9.11 within-day at 0.16, on one payer.

Hourly, country and aggregate

The period's motion: spend by hour, and impressions against installs

CTR, CPM, click-to-install and CPI by hour, T2 (Jul 23)

Spend against within-day revenue, by country

Cost per install by country

The three revenue measures on both periods, ROAS above each bar

Booked revenue by how long its payer had been installed

A. Per asset — delivery

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

asset	spend	impressions	CPM	clicks	CTR	installs	click→install	CPI
`T2a/control_sexy#2`	$28.18	1,303	21.63	121	9.29%	13	10.74%	$2.168
`T2a/utility_sexy#3`	$27.07	1,377	19.66	175	12.71%	22	12.57%	$1.230

The one place this month where two arms were matched on money and both delivered. utility clicked 12.71% against control's 9.29%, thirty-five installs behind it.

CTR by hour, one line per asset

Click-to-install by hour, one line per asset

Cost per install by hour, one line per asset

CPM by hour, one line per asset

B. Per asset — the three revenue measures

asset	booked	bROAS (Meta)	within-day	wROAS	cohort@H	cROAS	payers	payer%	ARPU	ARPPU	largest payer
`T2a/control_sexy#2`	$9.11	0.32 (0.32)	$9.11	0.32	$9.11	0.32	1	7.7%	0.701	9.11	100.0%
`T2a/utility_sexy#3`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—

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

One payer in the campaign, and it sits on the arm that delivered worse. Not readable at this sample size, in either direction: a delivery win of this size and a revenue loss of one person are two different questions, and only the first has a sample.

Payer rate by day, one line per asset

ARPU by day, one line per asset

ARPPU by day, one line per asset

Within-day ROAS by day, one line per asset

Composite: every asset across every measure, normalised across the period

C. Per asset — country mix

asset	spend	India %	US %	other %	US % of its revenue
`T2a/control_sexy#2`	$28.18	0.0%	0.0%	100.0%	0.0%
`T2a/utility_sexy#3`	$27.07	0.0%	0.0%	100.0%	—

Both rows are 100.0% other with no India and no United States, so the two arms bought the same geography and the delivery difference above is not a mix difference.

Where each asset bought: share of its own spend by region

Cost per install by region, every asset

Within-day revenue by region, every asset

T3 (Jul 23) — every section below this line is T3 (Jul 23)'s: the day, the hours, the countries, the campaigns and the assets, counted on T3 (Jul 23)'s own ledger and its own accounts. Use the buttons above to switch.

The first-round volume arm, buying India and South-East Asia. $40.68 for 5,902 impressions at a CPM of 6.89, the second cheapest in the month, 118 installs at $0.345 and $67.82 within-day at 1.67, on five payers.

Hourly, country and aggregate

The period's motion: spend by hour, and impressions against installs

CTR, CPM, click-to-install and CPI by hour, T3 (Jul 23)

Spend against within-day revenue, by country

Cost per install by country

The three revenue measures on both periods, ROAS above each bar

Booked revenue by how long its payer had been installed

A. Per asset — delivery

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

asset	spend	impressions	CPM	clicks	CTR	installs	click→install	CPI
`T3a/utility_sexy#2`	$28.90	4,173	6.93	478	11.45%	114	23.85%	$0.254
`T3a/control_sexy#3`	$11.78	1,729	6.81	114	6.59%	4	3.51%	$2.945

The same matched-arm shape as the 07-23 mid-tier round and the same direction, at larger counts. On impressions costing effectively the same, 6.93 against 6.81, utility clicked 11.45% against 6.59% and turned 23.85% of its clicks into installs against 3.51%, ending at $0.254 an install against $2.945. Spend was not matched here, $28.90 against $11.78, so read the rates and not the totals.

CTR by hour, one line per asset

Click-to-install by hour, one line per asset

Cost per install by hour, one line per asset

CPM by hour, one line per asset

B. Per asset — the three revenue measures

asset	booked	bROAS (Meta)	within-day	wROAS	cohort@H	cROAS	payers	payer%	ARPU	ARPPU	largest payer
`T3a/utility_sexy#2`	$93.57	3.24 (2.23)	$67.82	2.35	$71.38	2.47	5	3.7%	0.502	13.56	51.7%
`T3a/control_sexy#3`	$0.00	0.00 (0.00)	$0.00	0.00	$0.00	0.00	0	0.0%	0.000	—	—

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

2.35 on five payers, with 51.7% of it one person. Not readable at this sample size, and the campaign's 1.67 in section 5 is this same cell diluted by the arm that produced nothing.

Payer rate by day, one line per asset

ARPU by day, one line per asset

ARPPU by day, one line per asset

Within-day ROAS by day, one line per asset

Composite: every asset across every measure, normalised across the period

C. Per asset — country mix

asset	spend	India %	US %	other %	US % of its revenue
`T3a/utility_sexy#2`	$28.90	79.4%	0.0%	20.6%	0.0%
`T3a/control_sexy#3`	$11.78	57.8%	0.0%	42.2%	—

The two arms did not buy the same mix: 79.4% India against 57.8%, with the rest in the other T3 countries. India is the cheapest and best-converting of them, so part of the delivery gap above is where the auction placed each arm.

Where each asset bought: share of its own spend by region

Cost per install by region, every asset

Within-day revenue by region, every asset

7Caveats

⚠ The instrument gate is outside its band and is not rounded in. On the META clock, the only clock on which the check is valid, our own records read 1.07 of Meta's booked $3,242.34 against a documented 0.91 to 1.06. On the report's own UTC clock the same ratio is 1.08 and is not a valid check at all, because Meta's conversion value sits on its conversion hour and that is the one column an hourly export cannot restitch across two account days. Neither figure touches within-day, which never reads Meta's column; both bound any sentence that sets ours against theirs.

⚠ The payment repair does not reach July. The measure discards any payment it cannot tie to a person, and the 2026-08-16 repair recovers those by joining on an account UUID that the registration push began sending on 2026-08-13 18:52 UTC. July's cohort registered long before that, so the repair cannot reach it, and the payments the earlier feed could not attribute stay unattributable (asset_performance_report_2026-08-15.md). Whatever July is missing understates revenue, and the instrument gate reads above Meta, so the two do not compound.

The German buy's ch_ and de_ prefixes name a targeting split. The two sets share clips, captions and voice and differ only in the music bed (ad_ops/dach_stage2_asset_proposal_2026-07-30.md). DE/ch_arab_gulf_6s bought 498 impressions and one install at $21.750.

⚠ Nothing on this page is regime-matched. A regime is a stretch over which the ad set's daily budget did not change, and on this account the budget is a geography dial (DECISION_LOG.md #47). The estimator's hand-recorded regime tables cover the earner and the German buy only; the six tier campaigns have none at all. Inside these nine days the earner alone moved from $150 a day to $250, $500 and $875, and the 07-23 build's ad sets were raised from $5 to $20 with no timestamp on record. Every per-campaign and per-asset figure above therefore pools across budget changes.

⚠ The country section and every table C are the account's own days, seven hours later than the rest of the page. On the earner that window reaches into 2026-08-01 UTC and shows $408.99 and $150.33 where table A shows $312.93 and $8.64. Read them for mix, never for level.

The cohort here is complete, which no daily in this series can say. 5,069 installs whose ages at the payment cut run 371.0 to 567.3 hours, 100% covered at H=24, so $2,697.93 is frozen. Nothing in the cohort column can rise and no coverage caveat attaches to it.

The tier frame this month tested was retired inside the month. DECISION_LOG.md #36 dropped the T1/T2/T3 cut on 2026-07-26 because the tiers were assumed-cost groupings that came out inverted on measured paying rate and each bundle averaged incompatible members. The six tier campaigns are reported here because they spent.

The 2026-07-25 build could not answer its own question and this report inherits that. Eighteen cells, one ad set each, no replication anywhere, which is where the 16-payer bar came from (DECISION_LOG.md #37). Three campaigns of eight and five assets of the whole roster clear that bar in nine days of buying.

Two install counts run through this page and they are never crossed. Meta's 4,561 against our 5,069 registrations. Meta's is the delivery funnel's numerator and denominator; ours is the denominator under every payer rate and revenue measure.

The page carries 18 charts at account level and 18 more inside each of eight tabs. Every one is inlined as SVG at publish time, so a hidden tab still ships its charts and the page is large. Moving them out of the tabs would reintroduce the chart appendix this format deleted, so they stay.

8What would settle the open questions

Whether India at 1.00 is a business. It is the only cell in the month with enough payers behind it to carry a level, and it landed exactly on the money paid for it before any store cut. Nothing inside July can move that; the August report is the first that can say which way it went.

Whether any creative in July differed from any other on revenue. Five assets cleared 16 payers and four of them sit inside one campaign, none of them regime-matched. The account's own decidability arithmetic puts the requirement thousands of installs per arm above what a nine-day round of eighteen-cell tests produced, so the honest answer is that July did not test creatives, it tested geographies.

What the tier round actually cost. Section 5 prices the six tier campaigns individually against the earner and the German buy. That table is the complete accounting and it does not need another period to be read; what it cannot say is what the same money would have done inside a single worldwide buy, which is the shape the account moved to on 2026-07-30.

9What this hands to the next read

August is the first period in this series with something behind it, and this report is that something. What it hands over is a level: an account CPI of $0.538, a within-day return of 0.95, a payer rate that put 233 people behind $2,344.02, and a settled cohort figure. Those four are what an August comparison is drawn against.

Three things need carrying in the head while that comparison is made. The nine days are not nine days of the same account, because six of the eight campaigns were switched off or restructured inside them. The earner ran on two of the nine and quadrupled its budget while it did, so a July-to-August earner comparison is two days of ramp against a different ramp. And the country cut is on the account's own clock at both ends, so it stays a mix reading in August too.

10Reproduce

Report clock UTC. B1's exports are stamped in its own account zone, UTC-7, so every delivery row above is restitched from two Meta days. The country tables and every table C are not restitched and are labelled where they appear.

run	`_runs/2026-08-16_periods/`, generated by `generate.sh`
period	2026-07-23 → 2026-07-31 UTC, nine days, nothing before it
config	`_runs/2026-08-16_receipt_series/run_config_full.json`; the instrument gate on `run_config_full_meta.json`
payment cut	2026-08-16 10:59 UTC on the report, 2026-08-16 03:59 on the account's clock for the gate
Meta exports	hourly per Meta day from the run config's archive, all 24 buckets verified; day and country×day from `_runs/2026-08-16_2026-08-15_daily_auto/exports/`
registrations and payments	`_runs/2026-08-16_2026-08-15_daily_auto/lum`
FX	`_runs/2026-08-02_today_read/fx_rates.json`
regimes	no regime table covers the tier campaigns; the earner's is `REGIMES_260730` in `ad_ops/asset_estimator/cohorts.py`

What the account spent in July that this report does not count. From 2026-07-15 to 2026-07-21 it ran the youguqi_ HeiHa test, $477.52, for a different app on a borrowed attribution stack. On 2026-07-30 and 2026-07-31 it also ran 0730_starfall_, $71.62, 代投 for a third party's app. Neither is this app's money, both are marked as skipped in the run config, and together they are why the account's own delivery export shows more July spend than the $2,455.00 on this page.

CFG=_runs/2026-08-16_receipt_series/run_config_full.json
python -m ad_ops.measures      --config $CFG --from 2026-07-23 --to 2026-07-31
python -m ad_ops.report_tables  --config $CFG --from 2026-07-23 --to 2026-07-31
python -m ad_ops.report_tables  --config $CFG --from 2026-07-23 --to 2026-07-31 --campaign-table
python -m ad_ops.report_tables  --config $CFG --from 2026-07-23 --to 2026-07-31 --campaign t3_0725
python -m ad_ops.report_charts  --config $CFG --from 2026-07-23 --to 2026-07-31 \
    --out ad_ops/figures/month_2026-07 --md _runs/2026-08-16_periods/chartblock_month_2026-07.md \
    --relpath figures/month_2026-07
python -m ad_ops.measures --config _runs/2026-08-16_receipt_series/run_config_full_meta.json \
    --from 2026-07-23 --to 2026-07-31

The last command is the only one whose instrument line is a valid gate, because it runs on the account's own clock. generate.sh runs all six periods and every tab in one pass; each campaign is written to its own --out directory, since the chart filenames do not carry the campaign and two campaigns sharing a directory would overwrite each other.

11Appendix — definitions

The report period is a run of UTC days. Adjust reports in UTC, so a UTC-bucketed report lines up with the attribution dashboard without anyone converting in their head. B1 is fixed at UTC-7, so a UTC day on that account is Meta day D−1 17:00 to 23:59 plus Meta day D 00:00 to 16:59.

The hourly section pools hour of day across the period. Nine days hold 216 hours and the table has 24, so each row is nine hours of buying stacked on top of each other, one from each day. It is read for diurnal shape and never as a timeline, and no row in it is an event.

Level totals do not travel between periods of different length; rates do. This report prints one period and makes no comparison at all, so nothing here is at risk of it, but the rule is why the August report compares CPM, CTR, click→install, CPI, ROAS, payer rate, ARPU and ARPPU.

The three revenue measures, never one, never mixed, always labelled:

measure	counts	moves after the period closes?
booked	revenue that arrived inside the period, whatever day its payer installed	settles about four hours after midnight, then fixed. Meta's own number is this basis
within-day	installed on a day and paid before that day closed	never. Sealed at midnight
cohort at H	that day's installs, counted within H hours of each person's own install. House horizon H=24	rises until every install has lived H hours, then frozen. Always state coverage

Within-day is the day-over-day measure because it is immutable and needs no horizon, at the cost of censoring by hour of day: a 01:00 install has 23 hours to pay before midnight and a 23:00 install has one. Cohort at a fixed horizon is the hour-neutral one. Booked is what Meta will quote at you, so it is reported to explain the gap and never to rank. All three read our own payment records, and Meta can express only booked, so it rides bracketed beside that column and never as a row.

Over a period, booked splits by how long a payer had been installed. Over a single day, lag and install day are the same thing; over nine they come apart, and the zero-lag row of that split is exactly the within-day total.

Two install counts exist and they are never crossed. Meta's Leads and our own registrations disagree by 5 to 10%. Meta's is used for the delivery funnel, so CPI and click→install have both numerator and denominator on one instrument. Ours is used for 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 the 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 were compared on different audiences.

The 16-payer bar (DECISION_LOG.md #37): below roughly 16 payers a revenue figure is a direction and not a magnitude. Revenue per install on this account is a near-zero vector with rare large entries, so separating two creatives by a fifth needs thousands of installs per arm.

What the campaign names mean. T1, T2 and T3 are country groups the account bought as a set, fixed in DECISION_LOG.md #24 and built as T1 for the United States, Canada, the United Kingdom, Australia and New Zealand; T2 for the United Arab Emirates, Saudi Arabia, Malaysia, South Africa and Mexico; and T3 for India, Indonesia, Nigeria, Pakistan and the Philippines. Two rounds of them ran: the 2026-07-23 build, whose assets carry an a in their label prefix, and the 2026-07-25 build, which carries a b. The first round tested two message concepts, utility for turning a photo into video and control for a companion who moves how you want, on one fifteen-second cut each. The second tested those two messages across three lengths in each of the three groups, eighteen cells with one ad set apiece. sasian_ and latina_es_ name the South Asian and Spanish-language creative packs added into the second round's India and Mexico ad sets on 2026-07-26; the word after them is the attire in the clip and the number is its length. The earner is the worldwide buy that replaced the tier structure on 2026-07-30, and the German campaign is the DACH test that opened the same day.

Internal — noindex. Not for distribution outside the team.