Cling helps users automate repeated web tasks step by step
Cling: Supporting Cross-Page Coordination with Human-Directed, In-Situ, and Small-Step Automation
Human-Computer Interaction
Summary
Using many web pages to get a job done can be slow and involve lots of clicking and switching tabs. The authors created Cling, a tool that finds small parts of this work that can be automated without losing user control. It shows these options right in the page, so users can pick and adjust them as they work. Tests showed Cling cuts task time in half and lets users work with fewer clicks and tab changes.
What this means in practice
- •For knowledge workers: Integrate Cling to speed up multi-page information tasks by automating repetitive coordination with fewer clicks and tab switches.
- •For productivity software developers: Build tools that embed small-step, user-directed automation in web workflows to enhance user control during complex task execution.$Commercial implications: This paper enables creating interactive automation features that improve productivity tools for workers managing multiple web apps.
Authors
Zhicheng Huang, Yining Cao, Haijun Xia
Abstract
Web-based information work spans independently developed pages, with goals and coordination needs often emerging through browsing. Manual coordination requires repeated low-level actions, yet explicitly specifying operations for automation can interrupt engagement with information that shapes users' next steps. Our approach constructs reusable, schema-based page representations and identifies coordination actions through correspondences across schemas. We instantiate this approach in Cling, which surfaces coordination opportunities in situ for users to select, configure, and execute incrementally while directing the broader task. A technical evaluation demonstrates accurate identification of useful opportunities across diverse domains. A 12-participant study found that Cling reduced task durations by more than half compared with both manual interaction and an end-to-end AI agent, while requiring 61% fewer clicks and 80% fewer tab switches than manual interaction. Its small-step automation supported control and verification as actions unfolded, while accommodating intent as it emerged and evolved through information encountered on the page.