Identifying suitable project funding is often challenging: user interfaces on public funding platforms frequently complicate this search, posing usability challenges for diverse user groups. Project initiators are often overwhelmed by the need to navigate complex forms and filtering options to refine their search input. Innovation consultants, who perform such searches regularly, primarily need streamlined access to detailed information. This work presents an interaction design approach for a funding recommendation system aimed at addressing these disparate needs. We propose an interaction model centered around natural language input for project descriptions. This approach aims to minimize upfront complexity and is designed to allow for tailored output views based on user needs. Project initiators seeking orientation can receive a simplified summary that prioritizes intuitiveness; innovation consultants requiring in-depth analysis are presented with a detailed, tabular view including ranked recommendations and justifications, emphasizing eficiency and depth. Simplifying input and tailoring output in this manner aims to make funding recommendation tools more usable and efective for a wide range of users.
Social entrepreneurship and civic innovation are important approaches to support the development
of community-driven solutions to a broad range of challenges. While identifying suitable funding is
necessary for project success, the usability of existing search tools often hinders the process. For example,
the interfaces of online platforms provided by German public agencies require users to navigate complex
forms and understand funding jargon [
The usability gap underscores the need for improved interaction design. Such improvements are
particularly necessary given the challenges of limited standardization in domains like government and
citizen participation [
Based on the usability challenges identified, our interaction design focuses on two core principles: simplifying the input mechanism and tailoring the output presentation. Addressing these key user interaction points aims to create a more inclusive and eficient user experience by reducing initial cognitive load while providing appropriate levels of detail based on user needs and expertise. To lower the barrier to entry and enhance eficiency, our system moves away from complex forms and relies primarily on a single input field for natural language queries. Users describe their project idea, goals, and funding needs in their own words, using a conversational input rather than a rigid database query format. This approach benefits project initiators who may lack familiarity with specific funding criteria or terminology, allowing them to express their requirements intuitively. Simultaneously, it ofers innovation consultants a streamlined method for query formulation compared to navigating multiple dropdowns or checkboxes. The system is designed to implicitly incorporate relevant contextual user data—like location or organizational details, when available and appropriate—or to request specific clarifications contextually, thus maintaining a clean interface while still enabling accurate matching. The system presents results through multiple tailored views and components, rather than a single, static display. This approach recognizes that users have difering goals and require varying levels of detail and interaction. For instance, a concise, consolidated recommendation view is ofered for project initiators seeking quick orientation. For innovation consultants or others desiring deeper analysis, the system ofers a detailed tabular view. This view presents funding sources ranked by relevance and includes essential structured data—for example, deadline, location, and funding amount. Additionally, the view provides system-generated justifications, thus enhancing transparency and enabling critical assessment. In addition to these views, the output includes content designed to guide the user toward better outcomes in subsequent interactions; this includes clarifying questions, or suggestions for improving the project idea based on common funding requirements or identified gaps. By ofering these varied views and guidance, the multi-faceted output addresses the initial usability challenges, ensuring that the presentation complexity and available information align with diverse user needs and analysis goals.
This work was partially supported by financial and other means by the following research projects: DUCA (EU grant agreement 101086308), DiProLeA (German Federal Ministry of Education and Research, grant 02J19B120 f), ROBIN (Grant no: KON-23-039) at the Bavarian Research Institute for Digital Transformation, as well as our industrial partners in the FinComp project.
The authors have not employed any Generative AI tools in the preparation of this work.