to streamline the reporting process by 70%
Companies face complex CSRD reporting requirements, often struggling with scattered data, manual work, and tight deadlines - leading to delays and errors. The ultimate goal is to create an AI-powered tool that reduces reporting time by 70%, simplifies CSRD reporting by automating data handling, identifies gaps, and enables faster, more accurate compliance.
Screen analysis of competitors' apps
AI tools analysis to understand the interaction between users and AI
Creation of new user flow
Redesign of high-fidelity prototype in Figma
Frontend development in Lovable
Significantly reduced reporting time – Automated data handling and AI-assisted workflows cut CSRD reporting time by up to 70%, helping companies meet tight compliance deadlines with ease.
Improved accuracy and compliance – AI-driven gap analysis and validation tools reduced manual errors, ensuring higher data quality and alignment with CSRD requirements.
Streamlined collaboration – Role-based access and real-time chat function improved team efficiency and reduced back-and-forth communication, accelerating the review and approval process.
Process
During the research phase, my goal was to deeply understand the competitive landscape and how AI features are integrated into real-world applications.
I began with a screen analysis of competitors’ apps, studying their UI structures, design patterns, and functionality to uncover best practices and opportunities for improvement.
I also explored apps using AI-powered features to see how users interact with AI features in interface
Existing insights – User pain points and personas were already defined, allowing research to focus on actionable design opportunities.
After getting an idea of what content and features were needed, we created a map of the app structure.
After defining the sitemap and core features, we purchased a Material Design system as a base. I customized the components to fit the product needs and built a high-fidelity prototype in Figma. In the next step, I translated the prototype into a working UI by developing it directly in Lovable. prototype link
What went well
Thanks to an effective screen analysis of competitors' apps, I was able to quickly identify a user-friendly and intuitive UI solution. Early user testing confirmed that the concept was easy to understand and met users' expectations.
What could have been better
It took some time to learn how to write effective prompts in Lovable. In the beginning, unclear prompts often affected unintended code areas or components, which led to extra debugging and rework.
The integration of the LLM into Lovable wasn’t entirely smooth — it required multiple adjustments and slowed down the development process at times.