Generative Künstliche Intelligenz im Webdesign-Prozess
: Eine prozessorientierte Einzelfallstudie am Beispiel der Microlab GmbH

  • Elisabeth Jessica Schneider

    Student thesis: Master's Thesis

    Abstract

    The rapid proliferation of generative AI is transforming creative workflows in web design. Companies expect efficiency gains, especially in early concept phases, while concerns remain regarding output quality and team acceptance. Against this backdrop, the present study investigates how generative AI tools like Uizard can accelerate the design process without compromising design quality, using the MICROLAB GmbH as a case study. The focus is on how AI affects lead times and the number and depth of iteration loops across the phases sitemap, user flow, wireframing, and mockup. Comparable client projects—with and without AI support—were analyzed based on time logs from Projectfacts and versioned intermediate outputs. Findings reveal significant time savings in wireframing and mockup creation, enabled by automated sketch-to-wireframe conversion, rapid variant generation, and mobile ready prototypes. However, manual finetuning remains essential to ensure compliance with typography, layout logic, and component behavior according to corporate identity (CI) standards. The productive advantage lies in combining fast AI-generated baselines with designer refinement. Feedback and approval rounds remain key time drivers—regardless of tools used. AI is most effective when embedded in clearly defined design systems, supported by prompt libraries, and governed through structured review procedures. Despite limitations such as small sample size and onboarding effects, the analysis demonstrates that generative AI can measurably accelerate web design workflows—provided it is embedded in standardized processes, guided by clear quality controls, and aligned with a consistent CI framework. For MICROLAB, this suggests a pragmatic path: use AI early for variation and prototyping, but keep final layout decisions within a structured, CI-based review process.
    Date of Award2025
    Original languageGerman (Austria)
    Awarding Institution
    • Johannes Kepler University Linz
    SupervisorJohann Höller (Supervisor)

    Studyprogram

    • Digital Business Management

    Cite this

    '