The Problem
A SaaS startup's leadership team spent 4 hours every Monday morning manually pulling numbers from Stripe, Google Analytics, and HubSpot — copying them into a Google Slides deck and writing a summary. It was the same repetitive work, every single week.
The Automation
I built an n8n workflow that runs automatically every Monday at 8:00 AM and delivers a complete business intelligence report to the entire leadership team.
What Gets Pulled Automatically
- Stripe — Weekly revenue, new subscriptions, churn, MRR
- Google Analytics — Sessions, new users, top pages, conversion rate
- HubSpot — New leads, deals closed, pipeline value, sales activity
What AI Does With the Data
All data is sent to GPT-4o with a structured prompt that:
- Identifies the biggest wins and concerns of the week
- Compares current week vs previous week (% change)
- Writes an executive-level summary in plain English — no jargon
How It's Delivered
The final report is:
- Formatted into a Google Slides template automatically
- Emailed to the leadership team via Gmail with a direct link
- Also posted in a Slack channel with the AI summary
Results
⏱️ Time saved: 4 hours every Monday — that's 16+ hours/month
📊 Consistency: Reports are always ready by 8AM, without fail
🤖 AI summaries: Leadership team rated the AI summaries as "better than what we wrote manually"
📉 Human error: Eliminated — no more copy-paste mistakes in KPI slides
Tools Used
- n8n — Workflow automation and scheduling
- Stripe API — Revenue and subscription data
- Google Analytics API — Traffic and conversion data
- HubSpot API — Sales pipeline data
- GPT-4o — Executive summary generation
- Google Slides API — Report generation
- Gmail — Report delivery
- Slack — Team notification