Moving a sophisticated marketing machine from on-premise to the cloud
client context
An independent marketing and technology company, the client hosts a network of directories to help bring together students and schools. Across a stable of sites, potential students can discover information on education and careers in various fields.
The client’s in-house marketing and analytics teams also help schools meet enrollment goals via data-driven solutions. Besides providing resources on the web, their feature packed app is designed for goal setting, time management, and personal insights – giving students tools to achieve academic success..
business challenge
The client focused significant time and efforts on innovation, but not on maintenance of their application infrastructure. This resulted in longer delivery times for each release, higher fragility in the stack and, and explosive growth of staff and processes just to maintain the existing code.
Additionally, as all environments were hosted on-premise, large investments were made to setup and maintain hardware and infrastructure to support just two annual peak periods – meaning that the remainder of the year they were still bearing high costs for idle equipment that could not be re-purposed.
THE Broadpoint solution
“The team built and executed a plan to consolidate the data stores and migrate them to Azure SQL. As data is the client’s core offering, our team also implemented security gates to protect that data. ”
BroadPoint Cloud team used proven methodologies to complete a comprehensive analysis of intake code and data stores. After the analysis, the team built and executed a plan to consolidate the data stores and migrate them to Azure SQL. As data is the client's core offering, our team also implemented security gates to protect that data.
Once the data stores migration completed, the team then took on the task of simplifying and migrating the job processing system to Azure. In the first phase, the team set up a virtual machine to host the job processing engine. Tasks were merged and consolidated where possible, to reduce the potential points of failure by over a third. Other tasks were rewritten and then the entire system was instrumented, and performance monitors put in place to react to any kind of anomalous behavior – frequently and in an automated way.
The second phase broke the job process into discrete Azure functions that react to conditions rather than run on a schedule. This equipped the client with near real time data – a sharp contrast to the previous system, which could be as much as 72 hours behind. Finally, the BroadPoint team became the client’s ongoing outsourcing partner - taking the responsibility of system development and maintenance off their plate – so the client could now focus on their core product offering.
Key Outcomes
• Enhanced data security measures
• 33% reduction in points of failure
• Availability of real-time data, critical for success of a data focused company
• Outsourcing partnership for system development and maintenance