See llms.txt for all machine-readable content.

Back to Templates

Score Meta ad account health weekly with Claude and Gmail

Created by

Created by: Nocode Expert || iamrup
Nocode Expert

Last update

Last update 10 days ago

Categories

Share

Quick overview

This workflow pulls hourly performance from the Meta Marketing API for the last 30 days, calculates cost-per-result by hour, asks Anthropic Claude to recommend a dayparting schedule, and emails a formatted report via Gmail.

How it works

  1. Runs when you manually execute the workflow.
  2. Fetches account-level Meta Ads insights for a selected date preset, broken down by hour of day in the advertiser time zone.
  3. Calculates cost-per-result per hour, compares it to the overall benchmark, and identifies best and worst-performing hours above a minimum spend threshold.
  4. Sends the benchmark and ranked hourly performance data to Anthropic Claude to generate a JSON dayparting recommendation (cut hours, scale hours, and a brief plan).
  5. Converts the AI response into a plain-text “Meta Dayparting Report” with the benchmark CPA and recommended hours.
  6. Emails the report using Gmail.

Setup

  1. Set the required environment variables for the Meta Marketing API request: META_ACCESS_TOKEN, META_AD_ACCOUNT_ID, and optionally META_API_VERSION.
  2. Add your Anthropic API key as an environment variable named ANTHROPIC_API_KEY.
  3. Connect a Gmail account in the Gmail node and set the recipient, subject, and any other email options you want.
  4. Adjust the configuration values (DATE_PRESET, MIN_HOUR_SPEND, and RESULT_ACTIONS) to match the conversion events you optimize for.