🤖 AI Summary
This study introduces Plan Mode—a novel interaction paradigm—into spreadsheet environments, a prevalent end-user programming context, to investigate its efficacy in supporting iterative, non-rigidly technical workflows. The authors developed a spreadsheet agent prototype incorporating Plan Mode and evaluated it through a within-subjects user study with 24 participants, comparing it against a baseline tool without planning support. Although Plan Mode did not significantly improve task performance outcomes, it markedly reduced users’ post-hoc adjustment behaviors and received higher subjective ratings in fostering creativity and enhancing human-agent collaboration. These findings underscore Plan Mode’s distinctive value in enriching user perception and improving the quality of interactive experiences, even when objective task metrics remain unchanged.
📝 Abstract
Plan Modes have become standard features in agentic programming tools, allowing users to gain transparency and control by working with the agent to develop a plan before task execution. However, it remains unclear whether the benefits of this feature translate to end-user programming environments such as spreadsheets. Since spreadsheet programmers tend to work iteratively and care less about technical correctness, upfront planning may not fit into their workflows as easily. In this paper, we build a prototype of a Plan Mode for spreadsheet programming and evaluate it against a non-planning baseline through a within-subjects user study (N=24). We found that despite similar task outcomes with both tools, using Plan Mode led to a reduction in refinement and a better perception of the tool across dimensions of creativity support and human-machine collaboration. We discuss the implications of these results for the future design of Plan Modes, and for the broader role of human-AI planning in end-user programming.