Case Study | Global Fashion Retail

Spike has been working with a leading online global fashion brand since our formation with a team of consultants leading a number of performance initiatives.

The platform’s complexity, the tools and technology in use as well as the high volume of transactions supported means that our delivery experience there is extremely relevant to any organisation needing to deliver change at pace in a large transaction volume scenario.

Benefits:

• Clear strategy gave direction and focus
• Developer mindset changed
• Testing early in the lifecycle driving problems out early
• Production tools gave confidence that test environment and models were accurate
• Client can prove they have capacity to support peak trading periods like Black Friday

Background

Our client is transforming the global fashion retail sector by implementing a digital platform that will flexibly enable global growth and provide new and innovative ways to interact and transact with their customers.

Our Approach

Our scope of delivery has been comprehensive, through the support of the creation of the overall strategy we have helped set out the model for delivering performance testing early in the lifecycle across a number of individual components and channels, driving out performance issues well in advance of the integrated state, saving time in root cause diagnosis.

From there we delivered the detailed plan for performance testing for a major re-architecting, moving from the legacy platform to a microservices, Azure based architecture. This involved:

• Volumetric analysis of major events including Black Friday to guide the construction of workload models
• Scripting and execution using Microsoft Visual Studio Load Test, data and environment set up
• Use of AppDynamics for analysis as well as testing in a production Azure Cloud to help prove the environments’ profiles.

We have also delivered custom courses and coaching to train developers in the art of performance testing and leveraged their skills early in the sprints to drive out issues earlier and embed a performance testing mindset within and across teams.

Doing performance testing early in the lifecycle has facilitated dramatic tuning improvements to individual services’ throughput and API response times.

Conclusion

• Retail e-commerce platforms are complex, fast-moving and benefit from a clear performance strategy and dynamic workload model
• Using production monitoring tools enhances performance testing by making the model and environments more realistic
• Coaching developers in performance testing changes mindset and drives potentially expensive problems out early, saving time and cost

Download as a PDF

No contact details needed, click below to download a copy of our case study with our compliments. Please consider the environment and read online or print responsibly.