SEO application stabilized, and user churn reduced to possible minimum
An application for SEO and site performance optimization is intended for marketing agencies that work with multiple clients. The service did not support the load that resulted in repetitive failures leading to users churn. The task was to find the cause of the unstable operation and eliminate it.
It was initially clear that the problem laid in architecture. But it was difficult to collect complete information as several teams had worked on the project before.
We performed a site audit. The cause of the failure resided in the application architecture. As a result, the service addressed the client server so often that the hosting server recognized it as a DDoS attack. Within 2 weeks the Enterprise-level architect and the DevOps team set up the infrastructure, made changes to the code, and stabilized the system. Users stopped leaving the portal. Based on the results of the work, we continued cooperation with the client as a dedicated team.