Monitor your brand’s Google suggestions with Bright Data, GPT-5.6, and Slack

Description

Quick overview
This workflow runs daily to monitor Google Autocomplete, People Also Ask, and related searches for your brand via Bright Data, classifies new suggestions with keyword rules and GPT-5.6, stores history in an n8n Data Table, and routes alerts to Slack and comparison topics to Google Sheets.

How it works
Runs every morning on a schedule.
Builds a set of brand + intent probes, then fetches Google Autocomplete suggestions and a Google Search results JSON (People Also Ask and related searches) via Bright Data.
Normalizes and deduplicates the returned suggestions, keeping only those that contain your brand.
Loads previously tracked suggestions from the brand_suggestions n8n Data Table and identifies which suggestions are new since the last run.
Classifies new suggestions using keyword rules first, then sends only the unclear ones to OpenAI (GPT-5.6) to label them as damaging, competitive, or neutral.
Upserts all suggestions back into the brand_suggestions table with first/last seen dates, then routes only new items by verdict to post a digest to Slack (damaging), append comparison topics to Google Sheets (competitive), or record and take no further action.

Setup
Create a Bright Data Header Auth credential and set up a SERP zone, then enter the zone name in the workflow Settings.
Add an OpenAI credential (for the GPT-5.6 model) and a Slack credential, and set the target Slack channel in Settings.
Create an n8n Data Table named brand_suggestions with columns suggestion_key, suggestion, source, verdict, first_seen, and last_seen.
Connect a Google Sheets credential, create a spreadsheet with a Comparisons tab, and paste the spreadsheet URL into the Settings backlog_sheet field.
Update the Settings values for brand, brand_context, seeds, and region to match the brand and market you want to monitor.