to simplify Ecovadis questionnaire responses.
Problem Statement
Filling out the EcoVadis questionnaire is a major headache for users. This time-consuming process often leads to errors, frustration, and missed opportunities for better sustainability ratings.
Business Goals
Build an AI-powered tool that simplifies the entire EcoVadis submission process.
Enable automatic filling of the questionnaire using uploaded documents and AI matching.
Provide intelligent gap analysis to highlight missing answers or weak areas.
Generate strategic recommendations to help companies improve their sustainability performance and score.
Achieve up to 90% time savings by automating repetitive work and reducing manual effort.
Screen analysis of competitors' apps
Discovery calls with clients
Creation of new user flow
Redesign of high-fidelity prototype in Figma
Frontend development in Lovable
Companies saved up to 90% of the time previously spent on questionnaire completion.
Teams gained clear guidance on how to improve scores, based on AI-generated recommendations.
Manual errors were reduced significantly thanks to better structure, automation, and validation.
Process
During the research phase, my goal was to gain a clear understanding of the competitive landscape — especially other tools offering AI-supported questionnaire completion. I analyzed how these tools approached automation, user experience, and document handling to identify best practices and gaps in the market.
In parallel, our product team conducted discovery calls with clients to better understand their workflows, pain points, and expectations. These conversations revealed common frustrations such as time-consuming manual input, unclear documentation requirements, and the lack of guidance for improving EcoVadis scores.
These insights laid the foundation for prioritizing features like automatic document matching, gap analysis, and clear next-step recommendations.
To ensure the solution meets real user needs, we defined key personas based on research insights and discovery calls. These personas helped us understand different roles, goals, and pain points across teams involved in completing the EcoVadis questionnaire.
We quickly built the first version of the product using AI prompting in Lovable, based on initial requirements. This MVP was created within a week and served as the foundation for early user testing.
User feedback revealed that the original flow was difficult to understand. Test users expressed a strong preference for a structure similar to the official EcoVadis questionnaire, as they were already familiar with it.
Based on these insights, I analyzed the EcoVadis structure, extracted the most effective UI patterns, and redesigned the prototype in Figma. The new version was tested again and proved to be much more intuitive for users. After validation, we implemented the improved design in Lovable.
What went well
We kicked off the project with a very effective workshop led by our PO, which helped align stakeholders early and collect valuable input. The team also quickly organized discovery calls with our target user group, giving us fast access to real user insights. In addition, the Marketing team hosted a seminar focused on the EcoVadis topic, helping us understand pain points and business goals early in the process.
What could have been better
The first prototype was created directly in Lovable using AI prompting. While this allowed for rapid development, Lovable’s limitations around UX/UI best practices became clear during testing. As a result, we had to rebuild the entire user flow in Figma, which cost us about two week of extra time.