Kintra
Role
UI Designer, UX Researcher
Users
HR Admins, Team Managers, Employees
Tools
Figma, OpenAI, Informal Interview
Timeline
4 Weeks
(01) Overview
Context & Motivation
This project focuses on designing a leave management experience for small teams to SMEs—teams that need structure without the overhead of enterprise HR systems. The goal was to create a focused, easy-to-use solution that supports everyday leave workflows while remaining lightweight and scalable.
Problem
Many small teams manage leave through chat messages, spreadsheets, or email threads. As teams grow, this leads to missed approvals, unclear leave balances, and overlapping absences that are discovered too late. HR often ends up manually coordinating approvals and answering repetitive questions about leave status and policies.
Existing tools tend to be part of larger HR suites, introducing unnecessary complexity. This project aimed to solve a narrower problem: making leave requests, approvals, and visibility simple and reliable.
Challenges & Constraints
Access to users was limited to one HR interview
Scope limited strictly to leave management
Designed for three roles with different responsibilities
Avoiding feature bloat while still supporting real-world edge cases
(02) Research & Ideation
Leveraging a single user interview and secondary research
With limited user access, the interview strategy prioritized depth over volume. One in-depth interview with an HR practitioner focused on approval ownership, common delays, handling of unpaid leave, and visibility gaps across teams. Rather than exploring all HR workflows, questions were carefully chosen to surface high-impact leave-related pain points.
The interview revealed that HR often chases approvals, employees frequently ask about balances and request status, and overlapping leaves are usually discovered too late. These insights were reinforced through secondary research, including reviews of leave management tools and discussions around small-team HR workflows. Across sources, similar issues consistently appeared.
Notes from the user interview that shaped early assumptions and constraints.
Using heuristic evaluation
A heuristic evaluation of existing leave management products revealed recurring usability problems. Approval actions were often hidden under multiple steps, terminology around paid and unpaid leave lacked clarity, and dashboards attempted to show too much at once. Critical signals—such as pending approvals or conflicts—were frequently buried.
These findings directly influenced layout decisions, information hierarchy, and labeling across the design.
AI ideation
AI was used as a support tool during early ideation. It helped generate layout variations, navigation structures, and potential edge cases quickly. These outputs were treated as prompts rather than solutions and were filtered through heuristic principles, interview insights, and scope constraints before being considered for design.
(03) Design Phase
Design Approach
The design phase focused on intentional reduction. Rather than building a full HR system, features were chosen based on frequency and impact. Every screen and interaction was evaluated against a single question:
Does this help someone request, approve, or understand leave more easily?
The product was designed around three roles:
Employees, who need clarity and confidence when requesting leave
Managers / Approvers, who need speed and context to make decisions
HR, who need oversight without being pulled into every request
Role-based access ensures users only see what’s relevant to them. Users with full access can navigate Leave Management, My Leave, Leave Calendar, People, and Policies, while employees primarily interact with My Leave and the Calendar.
Leave Approval Flow
Approve or reject with full context at the point of decision.
The approval flow prioritizes speed and clarity. Managers can review leave details, remaining balances, and potential overlaps before approving or rejecting. Optional notes allow context for approvals, while rejections require a reason to ensure transparency and reduce confusion for employees.
Approved requests automatically move to the Approved state, reinforcing system feedback and reducing manual tracking.
Leave Request Flow
Review your leave details before submitting.
The leave request flow helps employees understand how their leave will be calculated before submission. Paid and unpaid days are clearly displayed, reducing misunderstandings around balances. Optional notes and document uploads support cases like sick leave, while immediate status confirmation sets expectations.
Main Screens
The main screens are structured around user roles to support everyday leave workflows without unnecessary complexity. Employees can view balances, submit requests, and track leave through My Leave, while the Dashboard surfaces key information such as pending actions and upcoming absences at a glance. Managers handle reviews through Leave Approvals, with enough context to make quick decisions, and rely on the Leave Calendar to anticipate team availability. For administrative tasks, HR uses Leave Management and Policies to define rules, accruals, and eligibility across different employee types, while People centralizes employee information to ensure leave decisions remain consistent and transparent across the organization.
(04) Outcomes
Key Learnings
This project highlighted the value of focus over completeness. Even with limited research input, combining a targeted interview, heuristic evaluation, and secondary research surfaced consistent and actionable insights. Designing for multiple roles did not require more features—only clearer boundaries around access and responsibility.
Visibility proved more impactful than automation. By surfacing pending approvals, overlaps, and upcoming leaves, the system reduces friction without adding complexity. AI was most effective when used to accelerate exploration, not decision-making.
Next Steps
Future iterations would include usability testing with managers and HR to validate prioritization and approval flows. Additional exploration could focus on scaling policies for more complex organizations and refining notifications based on real usage patterns.