Circus, Console Redesign
Transformed an internal web tool into a clear, task-oriented system that operators can finally trust.
Mar 1, 2025
Company
Circus Group
Company
Circus Group
Role
Designer & PM
Role
Designer & PM
Service
End-to-end Design
Service
End-to-end Design
Duration
5 weeks
Duration
5 weeks
Business Context
Business Context
In 2023, Circus acquired Aitme, a robotic kitchen company. Along with the acquisition came the Console, a web-based tool used to trigger tasks (like turning units on/off) and directly manage the robots. By 2025, we were ready to go to market, but the Console still looked and felt like an internal tool. It wasn’t scalable or suitable for B2B use.
Goal
Refine and redesign the Console to make it B2B-ready, ensuring a seamless operator experience and supporting a stronger, more credible sales narrative.
In 2023, Circus acquired Aitme, a robotic kitchen company. Along with the acquisition came the Console, a web-based tool used to trigger tasks (like turning units on/off) and directly manage the robots. By 2025, we were ready to go to market, but the Console still looked and felt like an internal tool. It wasn’t scalable or suitable for B2B use.
Goal
Refine and redesign the Console to make it B2B-ready, ensuring a seamless operator experience and supporting a stronger, more credible sales narrative.
Problem
Problem
Branding & Design System Misalignment
The Console, inherited through the Aitme acquisition in 2023, felt like a disconnected internal tool. Its interface didn’t reflect our design system, lacked visual consistency, and offered a fragmented experience across the product ecosystem.
Poor Fit for Operator Needs & Complex Architecture
Originally built for internal use, the Console was overloaded with technical features irrelevant to operators. Its structure was overly complex, navigation was slow and unclear, and key workflows were buried, making it unfit for scalable, external use.
Branding & Design System Misalignment
The Console, inherited through the Aitme acquisition in 2023, felt like a disconnected internal tool. Its interface didn’t reflect our design system, lacked visual consistency, and offered a fragmented experience across the product ecosystem.
Poor Fit for Operator Needs & Complex Architecture
Originally built for internal use, the Console was overloaded with technical features irrelevant to operators. Its structure was overly complex, navigation was slow and unclear, and key workflows were buried, making it unfit for scalable, external use.
Research Impact Highlights
Research Impact Highlights
Mapped the Full Operator Workflow
Shadowed 5 operators over a full shift to capture their end-to-end routine, from startup at 08:00 to shutdown at 19:00. Uncovered high multitasking and fragmentation across multiple tools, highlighting the need for better task consolidation.
Revealed Critical Gaps in Alert Handling
Once ingredients were loaded into the unit, operators moved on to other responsibilities (cleaning, procurement, processing orders). They needed reliable remote alerts, but instead, they were getting calls from engineers as a workaround. The system lacked proper multichannel monitoring and autonomy.
Matched UI Design to Task Complexity
Identified key shifts in workload: straightforward setup in the morning, intense multitasking during peak hours, and deep cleaning at close. This validated the need for task-specific UI modes and more flexible interactions, rather than rigid, one-size-fits-all screens.
Mapped the Full Operator Workflow
Shadowed 5 operators over a full shift to capture their end-to-end routine, from startup at 08:00 to shutdown at 19:00. Uncovered high multitasking and fragmentation across multiple tools, highlighting the need for better task consolidation.
Revealed Critical Gaps in Alert Handling
Once ingredients were loaded into the unit, operators moved on to other responsibilities (cleaning, procurement, processing orders). They needed reliable remote alerts, but instead, they were getting calls from engineers as a workaround. The system lacked proper multichannel monitoring and autonomy.
Matched UI Design to Task Complexity
Identified key shifts in workload: straightforward setup in the morning, intense multitasking during peak hours, and deep cleaning at close. This validated the need for task-specific UI modes and more flexible interactions, rather than rigid, one-size-fits-all screens.
Approach and Rationale
Approach and Rationale
Started with alignment workshops to clarify the Console’s target users and reduce stakeholder misalignment. Ran a heuristic audit to surface key issues in navigation, language, and information hierarchy. Organized a dedicated workshop with stakeholders to define how to simplify the Console’s architecture and make navigation task-focused rather than engineering-driven. Led a cross-functional ideation session with stakeholders and proxy users, centered around a core insight:
“During shifts, operators struggle to interact with units because the Console is complex and lacks actionable alerts. Critical tasks are buried, and operators rely on engineers to trigger actions—creating a bottleneck and making scaling impossible.”
From there, we collaboratively defined early directions: simplify the information architecture, surface key tasks, and reduce operator-engineer dependency.
Started with alignment workshops to clarify the Console’s target users and reduce stakeholder misalignment. Ran a heuristic audit to surface key issues in navigation, language, and information hierarchy. Organized a dedicated workshop with stakeholders to define how to simplify the Console’s architecture and make navigation task-focused rather than engineering-driven. Led a cross-functional ideation session with stakeholders and proxy users, centered around a core insight:
“During shifts, operators struggle to interact with units because the Console is complex and lacks actionable alerts. Critical tasks are buried, and operators rely on engineers to trigger actions—creating a bottleneck and making scaling impossible.”
From there, we collaboratively defined early directions: simplify the information architecture, surface key tasks, and reduce operator-engineer dependency.
Execution
Execution
Rebuilt the information architecture around operator goals and mental models.
Defined a new unit status logic model using clear language and icons.
Designed flexible UI modes for setup, monitoring, and troubleshooting.
Created prototypes and tested with internal stakeholders to validate flows early.
Rebuilt the information architecture around operator goals and mental models.
Defined a new unit status logic model using clear language and icons.
Designed flexible UI modes for setup, monitoring, and troubleshooting.
Created prototypes and tested with internal stakeholders to validate flows early.
Outcomes
Outcomes
Navigation now reflects operator workflows, no more digging through engineering tools to complete essential tasks. Statuses are now readable and actionable, helping operators respond faster and with more confidence.
Introduced a structured page system:
Home: real-time overview of robots and alerts
Unit: status, tasks, and actions per robot
Stock: ingredient levels and refill needs
Interactions: direct control (turn on/off, replace silos, etc.)
History: full log of past alerts and actions
Card-based layout supports multi-unit scaling, and the new IA improved cross-team alignment and reduced friction across the board.
Navigation now reflects operator workflows, no more digging through engineering tools to complete essential tasks. Statuses are now readable and actionable, helping operators respond faster and with more confidence.
Introduced a structured page system:
Home: real-time overview of robots and alerts
Unit: status, tasks, and actions per robot
Stock: ingredient levels and refill needs
Interactions: direct control (turn on/off, replace silos, etc.)
History: full log of past alerts and actions
Card-based layout supports multi-unit scaling, and the new IA improved cross-team alignment and reduced friction across the board.
Learnings
Learnings
Two weeks after launch, usability tests and surveys showed clear gains in efficiency, task speed, and user trust, highlighting the redesign’s impact and revealing the need for a mobile version to better manage alerts.
Two weeks after launch, usability tests and surveys showed clear gains in efficiency, task speed, and user trust, highlighting the redesign’s impact and revealing the need for a mobile version to better manage alerts.
