Designing AI Signals for Enterprise Operations

Overview

Zenda is a B2B enterprise platform that helps organizations understand how work actually gets done inside complex operations, not just how it is documented. The platform breaks work down into events, activities, and resources (people, tools, data, policies), making it possible to compare what a process is supposed to look like with how it is really executed on the ground. In this project, I designed a signal-based system that helps process owners quickly spot hidden cost issues and misalignments, turning complex operational data into clear, trustworthy signals while keeping decision-making firmly in human hands.

Design Process

1. Gather Requirements

   Understanding requirements and problem from the Product Owners and Leadership

2. Help Define Scope

   I had a say in defining the scope here since came up with a signalling system.

3. Desk Research

   Studying different signalling patterns in real world

4. Define Look & Feel

   Create options for final look & feel and present to design team and stakeholders

5. Create High Fidelity Mockups

   Creating final pixel perfect screens and pushing the new components to the design system

6. Review & Iterate

   Present mockups & proposed user flows to team and stakeholders, working quickly and iteratively

Problem Statement

Process owners and operational leaders were responsible for managing complex, evolving processes, but lacked timely and reliable ways to understand how those processes were behaving day to day.

Challenge 1: Limited visibility into where attention was most needed within large, interconnected processes. Challenge 2: Difficulty keeping documented process definitions aligned with real-world execution as teams, tools, and constraints changed over time. Challenge 3: Heavy dependence on manual reviews, static reports, and periodic audits to uncover issues.

“ How might we help process owners recognize what matters most within complex workflows, without overwhelming them or breaking the visual integrity of the system? ”

Users

Scope

1. 

   Activities with automation potential

   Steps that were highly manual, repetitive, or rule-based and strong candidates for automation.

2. 

   Cost outliers within the process

   Activities or variants that disproportionately drove operational cost or rework.

3. 

   Process mismatches

   Breakdowns between expected and actual process flows, including skipped steps, loops, or misaligned handoffs.

“ We needed a system of signalling to cut through operational noise and guide process owners toward what truly needed their attention. This system of signalling would also help if in case there are any future alerts needed. ”

Researching Existing Systems

I conducted an extensive research which led me to understand different signalling systems in the real world. I tried to decode what information they are trying to communicate through each pattern or color or shape selected. My exploration was around various systems like Heraldry, Nautical Flags, Japanese Mon, etc

Screenshot of my research on various signalling system

Design Decisions

Through desk research, I identified several recurring design principles used across different signalling systems. Based on these common patterns and similarities, I derived the core role signals should play within the product experience.

To design an effective system of signalling, I grouped signals under two primary sources:

Final Design

I conducted an extensive research which led me to understand different signalling systems in the real world. I tried to decode what information they are trying to communicate through each pattern or color or shape selected. My exploration was around various systems like Heraldry, Nautical Flags, Japanese Mon, etc

Prototype