When solving complex technical problems, having the right partner can make a big difference. Leveraging Amazon’s incredible technology offerings and access to their amazing engineering team via an AWS data lab helped us execute quickly and effectively.
How AWS Solved Our Big Problem In One Week
For many years iostudio carried technical debt on a particular data and analytics product for our biggest customer. The product was working fine to meet the contractual requirements, but we knew that our current offering was bloated, expensive and difficult to maintain. The modernization of the platform in question was the subject of much conversation between the Engineering team, Accounts managers and the customer.
The thought of spending thousands of human hours and dedicating a large contingency of money to address the problem was a hard pill to swallow, although there was agreement from all stakeholders that a decision needed to be made regarding if and when the modernization effort should take place.
Once a decision was made to proceed and all requirements were captured, we quickly learned that the proverbial shoe was dropping and our window for engagement was closing quickly due to a contract renewal period which had an impact on how and when we would proceed. At this point the Account Services team was asking (not really) for us to complete nine months’ worth of planned work and fit it into a three-month period.
The Solution
The Data platform in question involves infrastructure that collects data from a variety of internal systems, stores the information securely and then conducts a series of Extract, Transform and Load procedures before storing the end result inside a final datastore. The current platform provided a web-based output of information and more importantly gave vendors an ability to generate a customized PDF which had become the standard from our audience.
Many of the data schemas workflows did not require much change but our new requirement was to incorporate a real-time dashboard that could be embedded into the customers marketing portal which would provide users the ability to have much greater detail about all data points.
The iostudio Engineering team explored at least 6 commercial off the shelf solutions as well as the feasibility and cost of building a system from scratch. A member of the Data & Analytics team suggested a potential hybrid approach using AWS Services.
We were already leveraging AWS services such as EC2, RDS, S3, Route-53 and many others across our digital portfolio. Additionally, the AWS Gov Cloud, which already housed much of our infrastructure, operates many services within a FedRAMP space, which is a requirement for this solution.
All things considered, AWS was chosen as our solution. We use Step Functions to initiate processes, Lambda to house python scripts, Athena and AWS Glue to deliver data and Quicksight as the real time dashboard component.
How AWS Invested in Us
Our AWS Account Representative, who is very familiar with our business proposed an AWS Data Lab as a solution to our requirements. Although the pitch seemed a little “too good to be true” we agreed to a meeting because talk is cheap… right? Once we had supplied all our requirements to the data lab team – they laid out a plan which included a short series of solutioning sessions and a week long Engineering sprint. AWS brought the expertise and the time to methodically guide our internal Engineering team through the entire process of standing up a MVP which was ultimately very close to the finished product. Even post-launch the data lab team of experts had made themselves very accessible for follow-up, troubleshooting and even just general shoptalk.
To Sum It All Up
As AWS continues to build and support many of the most critical products and services that help make the digital world go around, their continued evangelism for client success is inspiring. The concierge level of service from the AWS Data Lab team was certainly the key to our success in building the right tool, the right way, with confidence.
Learn more about how we executed this in our AWS Data Lab Case Study.