Introduction
DipStudio created and maintains a once popular web application for posting and
sharing written content. When Scimus joined the application was still popular
and serving around 2 mln of users a month, but the technology stack they were
using got very outdated and was not able to handle all traffic coming, scaling
and new feature was an extraordinary task as changing anything on a website,
could cause a total collapse. At Scimus we proposed to update the technology
stack and migrate existing applications to a microservice based platform that
will be able to handle traffic better and be able to scale based on components
being loaded the most.
This approach allowed DipStudio to cut a time to
deliver new features in half, handle more traffic with less hardware, be able to
run different traffic experiments, and grow revenue due to less expenses on
maintenance and development.
Build software for advertising with our experts.
The Story of DipStudio
-
DipStudio is a company that has worked on the product since 1998 by building a community around it and serving approximately 2.5 million registered users. The company's main revenue streams were advertising and partnerships.
-
As new technologies emerged, the company realized that their outdated design and architectural approach needed to change to incorporate new functionalities without sacrificing quality.
-
The challenge was to modernize the product while maintaining its quality and keeping their primary audience in mind. The majority of their users were over 40 years old, and they were highly critical of new innovations. The company wanted to implement changes that would be comfortable for everyone and not disrupt the user experience.
-
This is where their collaboration with Scimus began, focusing on ensuring the product's quality and automating critical testing processes.
-
We took a user-centric approach and understood that every user mattered, which made the process even more challenging. We had to ensure that each user felt comfortable with the changes and that the product was compatible with the latest technologies.
-
It was a long and challenging journey, but in the end, we were able to successfully modernize DipStudio's product, introducing new features while ensuring that the changes were as seamless and painless as possible for all users. It was a rewarding experience, knowing that we had helped to breathe new life into a product that had been around for so long and had such a large community of users.
The Problem of Scalability
Due to old monolithic architecture and outdated resourceful technology stack, DipStudio was having issues with performance and was not able to serve all traffic at pick moments. Also a whole platform was running on dedicated servers and not in cloud which did not allow to easily scale application when traffic hits a server. This situation could be solved by buying more servers and investing more money in maintenance, but long term this solution would only cause more expenses and more resources to serve customers.
Why Scimus was chosen
DipStudio was looking for new resources to fill in a quality gap and found our profile on Upwork. Scimus agreed to provide test resources to test out all new changes that were done for an old architecture. At the same time we were looking for a way to speed up application and deliver best possible results.
How Scimus Responded
We started with quality assurance and automation of the testing process and after we became familiar with functionality, after that Scimus started to propose changes to architecture and a new look at what is being done right now. This proposal was supported by management of a company and we started to work on a new vision and were able to achieve significant results in a year. We are still working on this application, updating some modules and adding new features, but most of a heavy lifting is already done.
The Results
- We are able to rely on our history and long experience in vendor selection with work culture being a decisive factor, an attitude towards uneven workload distribution with it's highs and lows, experience and desire to learn, curiosity towards new approaches. Hard to boil it down to a formal selection process as we usually don't rely on one and choose with our guts.
- A fair amount of work was focused around improvement and optimization of web applications performance ratings (Lighthouse/Google WebVitals etc.). In over six years together we went through a handful of long integration-type projects, upgrading legacy systems with far more modern, mobile-first web API driven web apps. Backwards compatibility and seamless data migration was always a challenge along with high amounts of live traffic and user support requests that we were handling. Legacy technology stack had to be balanced with modern web application development frameworks (Node.js/React.js/TypeScript).
- At different times (with people coming and going) we have worked with over 15 people from Scimus crew in a range of positions, a minority being managerial and a majority in client/server side web development, mobile development, quality assurance (manual and automated testing) and technical specification.
