ROI-ai Search Experience Redesign
Transformed ROI-ai into a single high-efficiency sourcing tool by addressing fragmented data & inefficient searches, redesigned unified profiles, streamlined filters and automated workflows. Project duration was 4 weeks.
UX design roles
Evidence-based design, HMW strategy
Intelligent IA, scalable framework
Tools
Figma
Google Workspace
Overview
ROI-ai is an AI-driven recruitment automation platform designed to eliminate manual burdens like sourcing, screening and outreach. Its core value proposition lies in saving recruiters hours of effort and driving revenue through intelligent, automated workflows that replace tedious administrative tasks.
This project launched to create a system providing live data insights on both candidates and clients, syncing directly with a main database to ensure data integrity. By employing a user-centric design process, we aimed to solve data fragmentation and workflow friction, ultimately empowering recruiters with a high-efficiency sourcing tool.
Strategic Focus
We defined our project scope by targeting the most critical daily activity which is the search for candidates and clients. We recognized that a streamlined search experience was the most effective lever to drive adoption of our new data offering, making the recruiter’s workflow immediately more efficient.
To realize this vision of efficiency, we focused on four pillars: intuitive design for simple navigation, a unified profile to build data trust, intelligent AI-powered search for quality matching and streamlined workflows with one-click actions to minimize manual effort and maximize engagement.
Setting The Foundation
Our process began with a competitive analysis of four major platforms using Nielsen’s 10 Usability Principles as a benchmark. This allowed us to systematically identify market gaps, revealing that while competitors offered various features, they universally suffered from data fragmentation and search tools that often returned irrelevant results.
This external evaluation confirmed that the primary struggle for recruiters was not a lack of talent, but a lack of efficiency and data trust. By pinpointing where manual effort bottlenecked workflows, specifically in managing duplicates and tool-switching, we established an evidence-based foundation to solve the structural problems our users faced.
Defining The Problem
The core issue was defined by the real-world consequences of inefficient tools: recruiters were losing time to fragmented data, leading to delayed hires and lost commissions. These findings highlighted the urgent need for an intelligent platform that could unify data and reduce operational costs.
We framed our strategy using four “How Might We” questions to guide innovation. We asked how we might create a unified profile for complete data views, design AI-suggested filtering, streamline workflows with one-click actions and build a minimalist interface for immediate recognition-based interaction.
Research Insights
Research confirmed that the root cause of recruiter inefficiency was tool-juggling Recruiters were constantly switching between 2-3 platforms to manage data, causing significant context switching and manual update errors. This disconnect was the primary barrier to productivity.
Based on this, we defined our MVP by focusing on high-impact, low-effort opportunities. We prioritized a simplified search experience, a unified profile, and a centralized queue management system to rethink the entire experience from the initial search to final data synchronization.
Execution
Reorganizing search and unifying profiles to eliminate platform switching 
Non-intrusive design keeps data management transparent ensuring recruiter confidence 
Unified, expandable profiles with data toggles eliminate platform jumping 
Streamlining data-heavy tables with a minimalist, scan-first hierarchy
We reorganized the search flow with logical filter placement and a People/ Company toggle to maintain momentum. The primary card view was overhauled into an expandable unified profile, incorporating data source toggles to build trust and eliminate the need for jumping between multiple platforms.
For data integrity, we designed a non-intrusive queue management feature. This side-by-side view offers recruiters contextual control over pending syncs and merges, allowing them to choose between a full refresh or appending new data, ensuring the database remains clean without disrupting the sourcing flow.
Future Vision
To keep the MVP lean, we deferred several features focused on deep predictive analytics. Future recommendations include “smarter” unified profiles with relationship mapping and visual pipelines to track candidate progress, alongside folder-based organization for better job management.
We also proposed queue automation settings where the system can auto-sync data when confidence levels are high. This would allow recruiters to toggle automation on or off, maintaining manual oversight where needed while leveraging the full power of AI-driven data enrichment.
Takeaways
The project highlighted the highly personalized nature of recruitment workflows. During synthesis, we encountered contradicting preferences regarding specific workflow steps, which taught us the importance of designing for flexibility rather than a “one-size-fits-all” path.
In future iterations, I would conduct card sorting to further refine how recruiters prioritize filters. This would provide the quantitative backing needed to fine-tune the interface, ensuring that the most critical data points are always front-and-center for every unique user.
