AWS Cost Optimization Dashboard

🧾 Introduction

As businesses grow in the cloud, so do their bills — often without clear visibility. That’s what prompted this project: to build a custom AWS Cost Optimization Dashboard that would help a mid-size company better understand, manage, and reduce their cloud expenses.

Instead of relying solely on the AWS Cost Explorer console (which can be complex and limited in automation), I built a fully serverless dashboard that automatically gathers cost data, summarizes it, and presents it in a user-friendly visual format.

🎯 Project Goals

• Automate cost and usage data collection from AWS

• Track daily, weekly, and monthly costs by service and region

• Store reports in DynamoDB and S3 for querying and archival

• Present insights in a visual dashboard for finance and engineering teams

• Provide a foundation for tagging compliance and resource rightsizing
  

🔄 Data Collection

A scheduled CloudWatch Event (EventBridge rule) triggers a Lambda function daily at midnight.

The AWS Cost Explorer API is used to query:

• Daily spend

• Service-level breakdown

• Region-level costs

• Tag-level groupings (e.g., environment, project)

The Lambda function stores:

• Summaries in DynamoDB (for fast querying)

• Detailed reports in S3 as timestamped JSON files
  

🧠 Backend Logic: Lambda + Cost Explorer

The Python-based Lambda function:

• Assumes a billing-permitted IAM role

• Queries cost data for the past 24 hours

• Calculates:

  • Total daily cost

  • Cost per service

  • Top 5 regions by spend

• Saves summaries to DynamoDB

• Uploads full datasets to S3
  

📊 Frontend Dashboard

The dashboard is:

• Static (HTML/CSS/JS)

• Hosted on S3 + CloudFront

• Uses Chart.js for data visualization

• Loads and filters cost data client-side via AJAX

Users can:

• Filter costs by date or service

• View top-spending services or regions

• Track cost trends over time — without logging into the AWS console
  

🔐 Security Highlights

• Lambda IAM role scoped strictly to Cost Explorer and DynamoDB

• S3 bucket encrypted at rest with versioning

• CloudFront adds SSL/TLS and optional geo restrictions

• Access to the dashboard can be limited via signed URLs or IP restrictions
  

🏢 Real-World Impact

For the mid-size company I built this for:

• It gave finance and engineering shared visibility into daily AWS costs

• Exposed idle or underutilized resources

• Supported better tagging enforcement and chargeback

• Encouraged engineers to think in terms of cost per service, not just total bill

Now, the dashboard is reviewed weekly by engineering leads to identify savings.

🎓 Lessons Learned

• Gained hands-on experience with AWS Cost Explorer API

• Built a real-time, serverless analytics tool

• Transformed complex billing data into actionable insights

• Strengthened collaboration between DevOps and Finance teams
  

✅ Conclusion

This project gave the company real-time visibility into cloud spending and helped shift from reactive cost management to proactive optimization. It also showed how powerful AWS-native tools can be when combined effectively — no third-party tools required.

🛠️ Tech Stack

• AWS Lambda

• AWS Cost Explorer API

• Amazon DynamoDB

• Amazon S3

• Amazon CloudWatch

• Amazon EventBridge

• Amazon CloudFront

• Chart.js (frontend)