Quick overview
This workflow runs daily to fetch competitor web pages listed in Google Sheets, detects meaningful content changes using a text fingerprint, scores and summarizes changes with Google Gemini, and emails a ranked HTML digest via Gmail only when changes meet your significance threshold.
How it works
- Runs once per day on a schedule and loads the digest recipient, significance threshold, and diff size limits.
- Reads the list of tracked URLs and their last saved snapshots from a Google Sheets tab.
- Fetches each page’s HTML one at a time, converts it into normalized readable text, and generates a hash to detect changes while tolerating noisy markup differences.
- Skips AI analysis for unchanged pages, and for changed pages computes an added/removed line diff capped to the configured maximum.
- Sends the diff to Google Gemini to score significance (0–5) and generate a category, headline, summary, and “why it matters,” then merges the verdict back into the page record.
- Updates Google Sheets with the latest hash, snapshot text, and last-checked timestamp for every page.
- Aggregates the results, builds a sorted HTML email digest for changes at or above the threshold, and sends it via Gmail (or sends nothing if no notable changes are found).
Setup
- Create a Google Sheet with a tab named “Pages” and columns: url, name, last_hash, last_snapshot, last_checked, then paste in the competitor URLs you want to monitor.
- Add Google Sheets OAuth credentials and select your spreadsheet and the “Pages” tab in both the read and write Google Sheets steps.
- Add a Google Gemini (PaLM) API credential for the change-scoring step.
- Add a Gmail OAuth2 credential and set the recipient email address and thresholds in the monitoring configuration values.