
Designing a rule-based automation system for utility operations
Turned code-based detection logic into a no-code workflow for defining, testing, and monitoring operational rules.
0–1 Product Design | Complex Systems UX | Interaction Design

Project Overview
The challenge
How do you turn a technical workflow that required XML and SQL into a no-code, intuitive UI under a tight timeline?
What I designed
A no-code workflow automation that lets users to define detection logic, configure automated actions, test rules before activation, and monitor detections.
What changed
Delivered an MVP of a new automation capability
Established the foundation for self-serve automation and monitoring
My Role
As the sole designer, I led end-to-end product design in partnership with 1 PM and 2 Engineers.
Defined the user workflow and interaction model for complex detection rules
Partnered with engineering to resolve rule logic, validation, and edge cases
Worked with Product to define MVP scope under technical and timeline constraints
Documented interaction patterns and specifications to align implementation
Problem
The capability existed, but wasn’t usable in practice because:
Rule creation and maintenance required XML + SQL expertise, leaving day-to-day operation teams dependent on technical specialists to configure alerts.
Users had no way to preview a rule’s behavior before activating it. Without a clear way to validate outcomes, and false positives undermined user confidence and trust in the alerts.
Goal
Design a no-code workflow that enables non-technical users to define detection rules, test their behavior before activation, and monitor results over time.
Solution
Detection rules page

Define detection rules

Define conditions

Actions tab

Define actions

Test before activation

Monitor detections

Design Approach
How do I make system logic intuitive and trustworthy?
System logic should be manageable
Separated the workflow using tabs so users can follow through working in smaller chunks.
System logic should be visible
Used tree and card elements to help users understand the structure.
System logic must be testable
Introduced test functionality along with a place to monitor detection results.
Key Design Decisions
A. Making complex rule logic visible

A Detection Rule is created.
A Rule can have multiple Conditions
Conditions follow OR logic
A Condition is made of Condition Blocks
Condition Blocks follow AND logic
A detection rule could contain multiple conditions, with OR logic between conditions and AND logic within each condition’s blocks. I worked with engineering to understand this structure and translate it into a visual UI that users could inspect and edit.
To make the logic easier to understand, I used a tree structure for create and edit, and cards for read-only.
Tree - Create / Edit mode

Cards - Read mode

B. Making rules testable and observable
Users needed to understand what triggered a detection and verify how a rule behaved before and after activation.
I brought together pre-activation testing, contextual values in generated descriptions, and a centralized detection log to support validation and ongoing monitoring.

Process
I joined the project early, working with engineering to understand the underlying rule structure and define an interaction model that could be implemented within the MVP timeline.
I mapped the user journey into manageable steps, helping the team align on the end-to-end workflow and scope of the experience.


To reduce custom development, I reused established UI patterns where possible and collaborated with technical documentation on field definitions, validation, and help text.

This is an excerpt from the documentation that I owned as a reference.
This became the guideline for field validation as well as node-level feedback to help users understand potential errors during complex rule condition configuration and fix issues.

Applications
Public safety
Detect critical events (e.g., outages, gas leaks) in real time
Revenue protection
Identify failing meters using event patterns
Quality monitoring
Detect repeated threshold variances over time
Outcome
Delivered an MVP of a new rule-based automation capability
Established UI for defining detection rules, configuring actions, testing and monitoring detections
Created a foundation for self-serve automation in utility operations
As a new product capability, the MVP established the foundation for the workflow; usage and operational impact had not yet been quantified.