Find weak ads in a CSV and write new variants
A growth marketing team built a workflow that reads a CSV with hundreds of ads, spots the underperformers and writes replacements within strict character limits, using two specialised sub-agents.
Steps
- Export ad performance to CSV
- Give the character limits for headline and description
- Split the job: one sub-agent for headlines, one for descriptions
Starter prompt
Read ads.csv. Mark the bottom 20% by click-through rate, then write three new versions of each: headlines max 30 characters, descriptions max 90. Return a new CSV with old and new text side by side.
Guardrail: A human still picks which variants go live.
Prompts are our own examples. Results from sources are as reported by their authors.