Inspiration
Every organization pays for licenses like Slack, Zoom, or Microsoft Teams — even when many users don't use them regularly. We saw SaaS license waste as a real problem that costs startups and enterprises thousands of dollars every year.
We wanted to create a simple, serverless, AI-driven automation that tracks usage and sends alerts when licenses go unused — saving time and money.
What it does
Slack API is called by a Lambda function to fetch active users
DynamoDB tracks usage trends over time
Another Lambda function scans for inactive users (no activity for 4 weeks)
If users are inactive:
Sends a report via SNS email
Sends a Slack DM to the user asking if the license is still needed
Calculates estimated savings (e.g., 10 inactive × $8 = $80/month)
How we built it
We built LicenseShrinker AI, a serverless AWS Lambda system that:
⏱️ Automatically checks Slack user activity every 4 weeks
📊 Stores license usage history in DynamoDB
✉️ Sends smart reports via Amazon SNS (email)
🤖 Messages inactive users on Slack asking, “Do you still need this license?”
💰 Calculates estimated monthly savings
All built without any manual dashboard — just plug, deploy, and save!
Challenges we ran into
Setting up permissions for EventBridge Scheduler and Lambda Roles
Making Slack DM messages work via bot tokens
Handling first-run cases (empty database initially)
Setting up SES/SNS correctly to email reports
Accomplishments that we're proud of
✅ Fully automated a Slack license cleanup system using serverless AWS tools
🧠 Built a system that calculates real business savings (not just logs data)
🤖 Enabled direct user communication with a Slack bot that messages inactive users
💌 Delivered smart email reports without building a dashboard
🚀 Did it all with no manual inputs — 100% automated, smart, and scalable!
What we learned
How to integrate Slack API with AWS Lambda
IAM roles and permissions for DynamoDB and SNS
Using EventBridge Scheduler to trigger automation
Structuring a clean, serverless architecture for business use cases
What's next for LicenseShrinker AI
💡 Make it multi-SaaS: Add Zoom, Google Workspace, and Microsoft Teams integrations
📊 Add a Web Dashboard: So admins can review reports interactively
🔐 Use Secrets Manager instead of hardcoded tokens for better security
📦 Launch it as a plug-and-play tool for startups and enterprises
🧠 Add ML-based predictions: Auto-recommend who might become inactive soon
Built With
- api
- dynamodb
- eventbridge
- lambda
- python
- sns
Log in or sign up for Devpost to join the conversation.