Glassity cloud cost optimisation

Glassity could not calculate savings until a CTO or engineer granted access to AWS infrastructure and cost data. The interface had to earn that trust at the point of connection.

I designed the flow from registration to analysis. CloudFormation was the primary route, while CLI, IAM Console and Terraform remained available. Users could inspect the exact permissions before continuing.

I mapped the RDS decision, built a prototype and tested it in person with a senior software engineer and two co-founder CTOs.

• Participants expected an RDS change to require substantial engineering work.
• Technical terms and pricing were not always understood.
• The biggest percentage saving attracted attention before people considered the commitment or future demand.

Before making a long-term reservation, customers needed a clear baseline: what they paid today, what had already been optimised, how much coverage they had and where opportunity remained.

I grouped that information by database family and AWS account, kept historical cost and realised savings visible, and placed reservation and renewal actions beside the relevant context. The layout kept monitoring context and the next decision on the same screen.

Once a customer chose a reservation, the priority changed from speed to mistake prevention. The final step repeated the account, quantity, coverage, upfront and recurring costs, total saving and expiry.

It then required the customer to type “RESERVE RDS INSTANCES” before continuing. The added seconds were intentional: AWS Reserved DB Instances are non-cancellable one- or three-year commitments.