Atlas & Bone
Fashion / E-commerce← All clientsReport · May 2026
paid mediaNumbers from the portal, not typed. Fill the bracketed lines, drop the internal block, and it is the deck. The shape is fixed by the reporting SOP. Month 1 of the engagement.
Weekly message, this month to date (Monday, WhatsApp)
Hi team, quick October 2026 check-in from Atlas. Spend so far $1,474 over 17 days, 33 purchase at $45 each. [AM: one line on what changed this week and what we are watching.] We will send the full report at month end.
1. What happened
[AM: two sentences: the outcome in the client's words, and the one reason behind it]
Spend
$1,554
purchase
38
Cost per purchase
$41
Creatives live
0
0 angles, 0 copy variants shipped
Top ads
- AM - ASC - Tailored Tee 1$1,21430 purchase$40
- AM - ASC - Cuban 1$1955 purchase$39
- AM - ASC - Knit Polo 1$630 purchase—
- AM - ASC - Chino Raglan 1$591 purchase$59
- AM - ASC - Axis 1$242 purchase$12
2. What we are doing next
[Media buyer: the decisions made this month and why, from the decisions log, in three lines]
[AM: what the client will see change next month, matched to what they said they want]
· Nothing yet. No creative axis has two cuts thick enough to compare across 72 ad(s) in 180 days, which is the normal state at this spend — and "we do not know" is a better answer than a ranking of four conversions that somebody then builds a quarter around.
· SGD 13200 of spend sits on ads whose names say nothing about format, hook or angle, so it teaches us nothing. That is the price of the old naming, and it falls to zero on its own as new ads ship under the convention.
· The largest share of spend went to "AM - ASC - Tailored Tee 1" at 3.00x, while "AM - ASC - Axis 1" returned 12.53x on less. Worth asking rather than acting on: Meta may be funding the first because it contributes to conversions credited elsewhere, which last-click reporting cannot show.
· 1 ad(s) spent SGD 63 and returned nothing: "AM - ASC - Knit Polo 1". At this account's spend level that is a meaningful share of the month, and it is the cheapest thing on this page to fix.
· Month 1 of this engagement. There is little history to judge against yet, so the useful output is a baseline — cost per purchase, which formats reach people, and what a normal month looks like — rather than a trend claimed from two data points.
· No new angle was approved for May 2026 while the account was spending. Everything running is inherited, which is fine for a month and is how an account quietly ages.
3. What we cannot see this month
· Which of our angles produced which result. All 5 ad(s) this month predate the naming convention, and live ads are not renamed — so this is closed rather than outstanding. Ads created from now on carry an angle code and attribute themselves.
Internal, never sent
Effort: 5 ads live, 0 angles shipped, 0 formats. Retainer $1,500, spend is 1.0x the retainer.
- Landing page and checkout work 88% of people who added to cart never bought. The ads earned that intent and the site lost it, which is work we can quote for and which no amount of new creative fixes.
- A second funnel to test against Everything ran through one campaign ("AM - ASC"). On another account a lead-magnet funnel came in at a quarter of the CPM of a book-a-call funnel, and there is no way to know which suits this client without a second one in market.
- Creative refresh Nothing new was approved this month while the account spent SGD 1554. Every ad running is inherited, and inherited creative is the thing that quietly stops working.
The unfiltered version. If an account is going badly the AM must know before the client does.