Afra_Portfolio

Traditional UI to Conversation AI

Ask, Don't search!

Conversational AI transforms Employee Self Service from a navigation-heavy, form-driven experience into a guided, information-rich conversation that helps employees understand first and act with confidence

↓ 70%

Less Navigation

↑ 60%

Faster Task Completion

↓ 60%

Fewer HR Queries

CHALLENGE

Business Requirement

The organization aims to enhance the Employee Self-Service (ESS) experience by reducing the effort employees spend searching for information, navigating multiple modules, and understanding HR policies. Employees should be able to access information, complete requests, and receive guidance through a single conversational interface.

PROBLEM STATEMENT

The Problem with Traditional UI

A navigation-heavy, form-driven experience that creates friction and forces employees to hunt for answers.

Multi Page Navigation

Employees often need to move between different modules, policy pages, balance screens, and application forms to complete a single task

Policy Hunting

Critical information is usually buried within lengthy HR policy documents. Employees must search through multiple pages to understand eligibility, limits, documents before taking action.

High Cognitive Load

Users are required to remember information from different screens, interpret policy language, perform calculations, and make decisions while simultaneously completing forms, creating unnecessary mental effort.

HR Dependency

Due to uncertainty around policies, eligibility, and process requirements, employees often rely on HR teams for clarification, resulting in repetitive queries and reduced self-service effectiveness.

Understanding The User

Scenario: Ahmed wants to apply for leave encashment but is unfamiliar with the company's encashment policy.

Unsure which leave types are eligible for encashment.

Doesn't know how many leave days can be encashed.

Cannot easily understand how the payout amount is calculated.

Doesn't know how encashment will affect his remaining leave balance.

Has to search through lengthy policy documents for answers.

Must navigate multiple screens before completing the request.

Relies on HR for clarification before submitting the application.

DISCOVERY

Research & discovery

Identifying the opportunity through stakeholder discussions, workflow analysis, and user pain point mapping.

Methods

Stakeholder discussions

Understand business goals, operational challenges, expectations for AI-powered ESS.

Existing Workflow Analysis

Evaluate how employees currently complete HR tasks within the traditional ESS platform.

Competitor Analysis

Explore how modern HR platforms and AI assistants support employee self-service.

User Pain Point Mapping

Identify friction points employees face during HR transactions.

OPPORTUNITY

Defining the Opportunity

How might we help employees complete HR tasks without navigating multiple modules and reading lengthy policies?

Key Insights

A focused set of design challenges to elevate the experience from screens to conversations.

Employees spend more time finding information

Policies are difficult to understand.

Users don't know eligibility before applying.

Frequent HR dependency.

Achieve by

Reduce navigation effort

Minimize cognitive load

Provide contextual guidance

Reduce HR Depedency

Stratergy

AI UX Principles

Guiding the interaction between intelligence and users through clarity, context, and trust.

Principle 1
Intent
Recognition
Understands requests regardless of wording by mapping semantic meaning to system actions.
Principle 2
Contextual Awareness
Remembers conversation state and environmental variables to provide relevant assistance.
Principle 3
Progressive Disclosure
Reveals information step-by-step only when needed to minimize cognitive overload.
Principle 4
Personalized Responses
Uses employee-specific data and historical preferences to tailor every interaction.
Principle 5
Explainable Calculations
Uses employee-specific data and historical preferences to tailor every interaction.
Principle 6
Guided Conversations
Replaces complex forms with intuitive questions that adapt based on previous answers.
Principle 7
Error
Recovery
Handles invalid requests gracefully by suggesting alternatives or asking clarifying questions.

Design

Conversation Design Framework

A structured flow that turns a request into a guided, information-rich conversation.

Intent Recognition

Context Gathering

Policy Guidance

Recommendation

Confirmation

Submission

FLOW DIAGRAM

User Flow

A clean, modern flow that reduces friction and keeps the employee moving forward.

UI Challenge

Challenge in Conversational UI

A focused set of design challenges to elevate the experience from screens to conversations.

Shifting Screens to Conversations
Shifting from navigation-driven workflows to conversation-driven experiences.
Challenge:
Deciding what the AI should say, when it should ask questions, and how much information to reveal at each step.
Preventing Accidental Actions
A conversational interface can feel fast and automatic.
Challenge:
Ensuring that important actions such as leave encashment, loan requests, or reimbursements always required a clear review and confirmation step before submission.
Balancing Chat with Structured Inputs
Pure text input is not ideal for every task.
Challenge:
Finding the right mix of conversational interaction and traditional controls such as date pickers, steppers, file uploads, and confirmation cards.
Designing for Uncertainty
Employees often ask vague questions such as "Can I know my leave balance?"
Challenge:
Designing responses that clarified eligibility, explained next steps, and guided users without overwhelming them
Progressive Disclosure
Showing all information at once would make the conversation feel cluttered.
Challenge:
Surfacing policy details, eligibility, balances, and calculations only when they became relevant.