The retainer, on the record
A 30-person performance marketing agency running Meta and Google ads for consumer brands couldn't show a new client CMO which monthly changes a person actually decided, only a 14-slide screenshot deck.
Proving what a marketing retainer delivers each month
Connectors read ad platform exports and return every figure with its source rows. Skills write a cited creative brief, draft brand-checked copy variants and flag landing pages that break an ad's promise. Every spend change is a proposal a named buyer must approve, and a month-end ledger builds itself from the action log.
- all 6 approved spend changes carry a named approver, and the gate sent 1 approval back for a missing reason
- the brief's 9 lines carry 14 citations, all re-read from disk and re-summed with 0 failures
- Meta ad exports
- Google ad exports
- Connector tool functions
- AI skills
- Approval gate
- Action log
- Node.js
Same problem?Send it over.
Tell me what your version looks like and which tools it runs in. I reply with how I would build it.
-
Media and creative agency
The Monday margin watch that reads jobs, not people
1,124 timesheet rows read, 31 person ids dropped at load, 0 found in any output file
-
Brand agency
The coordinator job we rebuilt before it was posted
7 of the 9 bullets on the drafted coordinator posting now run as fixed workflows; 2 stay with people
-
Marketing agency
The clip desk that reads its own drop-off
the desk pinned 6 of 6 drops to a transcript line and an export row; against the drops we designed into the curves ourselves, it found 6 of 6 and 0 of 1 decoy dips fired