The project
Hunter Douglas, an expert in innovative window treatments, strives to be anything but traditional in how they respond to their customers’ evolving needs.
As part of its mission to deliver a complete and exceptional customer experience, they developed an IoT application for window blinds and coverings - which needed to scale and uphold the highest standards of Hunter Douglas.
Description
Hunter Douglas faced significant challenges transitioning their mobile app app from an MVP to a fully scalable product. They faced significant challenges in maintaining the highest standards for data retention, including ensuring consistent synchronization between the app and the backend, delivering predictable backend performance, and implementing robust error handling. All these efforts were aimed at guaranteeing an uncompromised, top-tier user experience. The state of the backend further compounded these challenges, making it difficult to build upon the existing solution or engage in effective project planning geared toward growth.
These difficulties challenges highlighted the urgency for a comprehensive solution, so they sought an experienced partner capable of not only addressing the existing issues but also delivering an innovative and high-performing application all while optimizing the overall development process.
Drawn to our expertise and extensive experience in rescuing and revitalizing struggling projects, they chose IBORN as their tech partner.
We had to start from the beginning, which always means gaining full understanding of the solution, where we are and where we want to go. We immediately saw that we are dealing with several dimensions of complexity that had to be addressed properly:
-
One solution for 12 different brands
-
Mobile apps on both iOS and Android
-
A large variety of different smart blind motors
-
A growing number of users in various countries and languages
-
An ambitious plan for new features
Therefore, the heart of the project involved migrating the backend from Swift's Vapor Framework, which proved unsuitable for the complexity and scale of the application, to a .Net core API, a robust, mature framework for building scalable, cross-platform APIs.
More techy details
Argumenting a switch of technology must be objective and backed by facts, in order to guarantee future-proof decisions.
Vapor, while performant, is not as mature for server-side applications, especially for large-scale or long-term projects. Struggling with maintainability and encountering limitations in the Swift server-side ecosystem, these issues might persist as the project grows.
Therefore we went with .NET Core, as it is known for excellent maintainability features. Its vast ecosystem, built-in support for dependency injection, middleware, and excellent development tools, make it easier to manage and scale APIs. The NuGet package manager offers many libraries for common needs (security, IoT integration, etc.).
Vapor's ORM, Fluent, is relatively young and might lack the robustness or maturity of ORMs available in other ecosystems like Entity Framework in .NET. Since database migrations are a crucial part of scaling a solution, we chose Entity Framework Core, the primary ORM for .NET. EF Core is very powerful with robust migrations, strong type safety, and tooling for database management. It’s generally a more mature option for database-heavy applications, especially those needing high reliability.
The goals
-
Establish a stable and scalable backend environment;
-
Maintain high-quality standards Hunter Douglas strives for;
-
Normalize and transform data for improved efficiency and easier management;
-
Develop a technical roadmap that aligns with the growth of the business;
-
Implement proactive monitoring and automated testing to ensure reliability.
The challenge
-
A backend that wasn't fit for growth handling multiple tenants and platforms;
-
Improving the app's quality to enhance the end-user experience;
-
Adding centralized logging, monitoring, analytics, and over 4800 automated tests for pre-release validation.
-
Undocumented code that required comprehensive refactoring;
-
Database (MySQL) normalization and improvement in order to provide integrity, performance, and to serve as an enabler for BI;
-
Gradually migrating users in phases, while supporting both flows.
The team
-
2 Fullstack Software Engineers
-
2 iOS Engineers
-
2 Android Engineers
-
1 QA Engineer
- 1 UI/UX Designer
Technologies
-
iOS
-
Android
-
MySQL
-
AWS
-
.NET Core
The result
The backend migration to a .NET Core API and data structure optimization delivered remarkable results:
-
Improved user satisfaction: A reliable, seamless app experience with zero downtime and no data loss or synchronization errors.
-
Scalability: A robust environment capable of supporting growth while minimizing costs during low-usage periods.
-
Future-ready architecture: A solid foundation for integrating new features, and GEN2 motors, into the app. Now the app is handling 20+ different types of motors.
-
Enhanced development process: Refactored code, detailed documentation, and automated testing streamlined ongoing maintenance and future project execution.
To further improve the system, we developed a solution that allows for easy integration with Power BI, empowering the client to make data-driven decisions. With centralized logging, monitoring, and analytics, coupled with over 4,800 automated tests executed before each version release, the system is now proactively monitored and releases are highly reliable.
This comprehensive approach has enabled their IoT mobile app to scale in line with the business growth of Hunter Douglas, ensuring it’s equipped to support their evolving needs well into the future. . With a future-proof solution in place, the focus can now shift from maintenance to innovation, driving long-term success.
Let's discuss your project.
We're all ears.
Drop us a line