← All work About Author
HR
UX Writing · HR Mobile App · April–May 2024

The App Wasn't
Broken. The
Language Was.

How a microcopy overhaul — and a set of annotated redesigns — transformed a confusing HR payroll app into a system users could actually understand.

Product
HR365 · TalentPro
Role
UX Writer + UX Critique
Scope
Microcopy · Visualization · Nav · Tone
Timeline
Apr – May 2024
TL;DR — for the recruiter who has 30 seconds
The short version of this case study.
Jump to → The problem What I shipped Impact Reflection
01
Overview
What the project was

Context

A functional app that users couldn't navigate.

HR365 (TalentPro) manages employee salary, attendance, and HR operations for corporate users. The features worked. The language around them didn't. Users were lost — not because the product was broken, but because it never told them what anything meant.

Client
HR365 · TalentPro Mobile & Tablet
My Role
UX Writer + Design Critique
Scope
Microcopy, Salary Visualization, Navigation, Tone Redesign
Goal
Make the system explain itself without adding more UI
02
Reality Check
What users were actually saying

User Signals

Real frustration is the best brief.

These weren't hypothetical pain points. They were direct signals from users trying — and failing — to understand their own financial data.

"

Salary details are confusing. I can't tell what I actually received versus what was deducted.

Salary Clarity
"

Hard to understand deductions. The numbers are there but I don't know what they mean or why.

Deduction Visibility
"

Navigation is not clear. Too many options — no idea which one I actually need.

Navigation Overload
The pattern: None of these complaints mention a missing feature. They mention a missing explanation. Users weren't asking for more — they were asking for clarity on what was already there.
03
Problem
Four systemic failures

Diagnosis

This wasn't a design problem. It was a communication failure.

Each issue operated at system level — not screen level. Fixing one wouldn't fix the experience. All four needed to shift.

Problem 01 · Microcopy
Labels without meaning.
CTAs like “Submit”, “Proceed”, and “Details” gave users zero information about what would happen next. They required users to guess — which meant hesitation, re-reading, or abandonment.
Problem 02 · Visualization (Critical)
A pie chart that asked users to do the math.
Salary shown in a donut chart with overlapping labels, no legend context, and colors with no semantic meaning. Users couldn't answer: “Where is my money going?”
Problem 03 · Navigation
Feature density without hierarchy.
7+ services shown at equal visual weight. The daily action (Clock In) competed with social features used once a year. No grouping, no priority signal.
Problem 04 · Tone
A transactional machine, not a helpful system.
When users encounter financial data, anxiety is the default. The copy offered no reassurance, no confirmation, no context — at exactly the moments people needed support most.

04 — Core Insight

Users weren't struggling because
the system was complex.
They struggled because the system
didn't explain itself.

Every piece of data was present. Every feature existed. But none of it communicated. The gap was not functionality — it was language.

05
Strategy
Four shifts that drove every decision

Approach

Every word on every screen was governed by one of these reframes.

Shift 01
Technical Language
Human Language
Shift 02
Data Display
Data Understanding
Shift 03
Actions
Guided Actions
Shift 04
System Output
Conversation
06
Execution
Screen by screen — problem annotated, solution designed

Execution · 01 — Salary Visualization (Tablet)

The donut chart that forced users to do the interpretation work themselves.

This was the #1 user complaint. The screen showed data accurately — but made users mentally decode it before they could understand anything.

⚠ Problem — Actual Screen, Annotated

Five specific failures mapped onto the live screen.

Red markers on the real TalentPro tablet screen. Each one points to a piece of UX that asks users to do work the system should have done for them.

⚠ Annotated Problem
Tablet salary earnings tab — annotated problems
1
2
3
4
5
  1. Section tab says “Earnings” — describes the container, not the answer. Users still don't know their take-home.
  2. Donut chart with overlapping 9.4% and 12.5% labels — physically unreadable. 4 colors carry no semantic meaning.
  3. No total salary anywhere on screen — users have to mentally add ₹20k + ₹4k + ₹3k + ₹5k to find their income.
  4. Title bar just says “Salary Summary” — no orientation, no month, no greeting, no context.
  5. Line-item list is correct but buried — users must scroll past the chart to reach the data they actually need.
✓ Proposed Fix — Redesigned Screen

Remove the chart. Surface the total. Let each line speak for itself.

Hi-fidelity mockup of the redesigned screen. Same data, reorganised so the system answers the question users actually bring to this screen: “What did I earn this month?”

✓ Proposed Redesign
11:28July 2023 · 2023-2024
Your Salary Breakdown
July 2023 · Paid on 1 Aug
3
Here's your summary, Rohan. Everything you earned and what was deducted this month.
Net pay — what you received
₹ 32,000
Credited to A/C ending 4471 · 01 Aug 2023
Summary
Earnings
Deductions
Earnings · 4 components
Basic PayCore fixed salary
₹ 20,000
House Rent AllowanceHRA · tax-exempt up to limit
₹ 4,000
Dearness AllowanceDA · cost-of-living adjustment
₹ 3,000
Special AllowanceVariable monthly component
₹ 5,000
Deductions · view full breakdown →
Total deductionsPF, Tax, Insurance combined
– ₹ 0
  1. Screen renamed to “Your Salary Breakdown” — names the answer, not the chart container.
  2. Net pay ₹32,000 pinned at top in one large number — users get the answer in under a second.
  3. Donut chart removed entirely — replaced with a clean linear list with a context hint under every row.
  4. Single greeting line provides orientation without adding extra UI elements.
  5. Deductions grouped below with a separate red-dotted section label — one scroll, no mental sorting.

Redesign mockup built in HTML/CSS · not a real product screenshot

Execution · 02 — Dashboard Navigation (Tablet)

The dashboard that showed everything and guided nothing.

7 features. No priority. No grouping. The most time-critical daily action — Clock In — competed visually with Birthday Buddies and Wedding Anniversary at identical weight.

⚠ Problem — Actual Dashboard, Annotated

Equal visual weight means zero hierarchy.

Markers on the real screen show exactly where the hierarchy breaks down — from the decorative welcome panel to the undifferentiated service grid.

⚠ Annotated Problem
Tablet dashboard annotated with 5 problem markers
1
2
3
4
5
  1. “Welcome Rohan” identity card occupies ~20% of horizontal space — enables no action, purely decorative.
  2. Clock In & Clock Out controls give no confirmation state after tap — users don't know if it registered.
  3. “Remark if any” placeholder — vague. Users don't know what this field is for or who reads it.
  4. 7 service tiles at identical visual weight — Leave Balance and Birthday Buddies look equally important. No grouping, no hierarchy.
  5. Bottom nav labels are fine, but above the nav users see no sequence of what to do first.
✓ Proposed Fix — Priority Tiers

Group by frequency. Lead with what users need every single day.

Three visual tiers — Tier 1 (daily critical), Tier 2 (weekly), Tier 3 (social). Each one looks different. Users can tell at a glance what's urgent and what's optional.

✓ Proposed Redesign
TalentPro · Dashboard26 Oct Saturday · 11:28 AM
Good morning, Rohan
You haven't clocked in yet today.
3
Tier 1 Do this first · Daily
WAITING FOR CLOCK IN
Start your day in one tap
We'll confirm location and time automatically.
Clock In →
Tier 2 Your work week
Leave Balance
14 days left
My Team Attendance
3 absent today
My Holidays
Next: 2 Nov
Tier 3 Social & updates
Announcements
Birthday Buddies
Work Anniv.
Wedding Anniv.
  1. Greeting replaces decorative welcome card — “Good morning, Rohan · You haven't clocked in yet today” turns empty space into orientation.
  2. Tier 1 — Daily, full-width, green: Clock In becomes the one unmissable action with a visible “waiting” state.
  3. Tier 2 — Weekly, medium weight: Leave Balance, Team Attendance, Holidays — each tile carries a live data hint.
  4. Tier 3 — Social, de-emphasised pills: Birthday, Anniversaries, Announcements — visually lighter, lower in the page.
  5. Remark field renamed “Add attendance note (optional)” — context + permission inside the attendance flow itself.

Redesign mockup built in HTML/CSS · not a real product screenshot

Execution · 03 — Side Menu Panel (Tablet)

A profile card buried under system controls — with destructive actions hiding in plain sight.

The slide-out menu mixed personal identity, navigation, data utilities, and settings at the same visual level. No section separation. No warning language on irreversible actions.

⚠ Problem — Actual Menu, Annotated

Three categories, zero separation, two dangerous actions.

Every row in this menu has the same visual weight — whether it opens a safe profile page or wipes attendance data. Users have no signal that some taps are consequential.

⚠ Annotated Problem
Side menu tablet annotated — danger zone issues
1
2
3
4
5
  1. Profile card (photo, Employee ID, designation) is decorative — shows data but enables no action or status.
  2. “Hide all seen events” — what events? “Hide” sounds like deletion. Users avoid unclear system actions.
  3. “Clear Attendance” — sounds permanently destructive. No warning in the label, no visible confirmation prompt.
  4. Language & Settings mixed with Dashboard & Profile at identical weight — navigation and system utilities are indistinguishable.
  5. Sign Out button styled prominently — good visibility, but placed beside destructive actions with no visual boundary.
✓ Proposed Fix — Three Sections + Danger Zone

Navigate · Preferences · Account Actions — each with its own zone.

The same features, grouped by risk and intent. The destructive zone is visually distinct and the labels include what actually happens before the user taps.

✓ Proposed Redesign
  1. Profile row carries live status — clock-in time and leave balance turn decoration into information.
  2. Navigate section — Dashboard + My Profile; always safe to tap.
  3. Preferences section — “Mark all notifications as read” replaces “Hide all seen events”, with a reassurance line.
  4. Red-bordered danger zone — “Reset attendance cache” + “Sign out” physically separated with warning copy.
  5. Destructive action labels include their own safety hint: “You'll be asked to confirm before anything is cleared.”

Redesign mockup built in HTML/CSS · not a real product screenshot

Execution · 04 — Mobile Dashboard Navigation

Same navigation failures. Smaller screen. Zero margin for confusion.

On tablet, feature overload is frustrating. On mobile, it causes task abandonment. The phone version compounds every dashboard hierarchy problem into a single viewport with no room to recover.

⚠ Problem — Actual Mobile Screen, Annotated

A 3×3 flat grid on the device people use most.

The real mobile dashboard — annotated. Notice the attendance card with no confirmation state, the missing greeting, and the flat service grid where social and functional features collide.

⚠ Annotated Problem
Mobile dashboard annotated — 5 problem markers
1
2
3
4
5
  1. TalentPro logo — fine, but the header carries no greeting, no current status, no context for the user.
  2. Clock In / Clock Out — no confirmation text, no acknowledgement, no visible state after tap.
  3. “View full month tracking” — critical feature buried below the card. Easily missed on first use.
  4. Flat 3×3 service grid — Birthday Buddies and Leave Balance are visually identical. No hierarchy on the device where it matters most.
  5. Bottom nav is fine — but the screen above offers no visual answer to “what should I do first this morning?”
✓ Proposed Fix — Mobile Priority Tiers

Three tiers, mobile-compressed. Clock In gets the hero treatment.

The same priority system as the tablet redesign, compressed for a phone viewport. The morning clock-in action becomes unmissable. Social features shrink into a pill row.

11:2844% ▣
Good morning, Rohan
Sat, 26 Oct · Clock in to start
3
T1 Today
TAP TO CLOCK IN
Start your day
Location will be shared securely
Clock In →
T2 This week
Leave
14 left
Team
3 out
Holidays
2 Nov
T3 Social & updates
◉ Announcements
◉ Birthdays
◉ Work Anniv.
◉ Wedding Anniv.
Dashboard
Salary
Tax
Inbox
Menu
  1. Header shows greeting + day + status — “Good morning, Rohan · Sat, 26 Oct · Clock in to start” replaces empty logo space.
  2. Tier 1 green card makes Clock In the single most visible action on the screen.
  3. Tier 2 tiles carry live hints: “14 left”, “3 out”, “2 Nov” — users don't have to open each feature to check basic status.
  4. Tier 3 pills for Announcements, Birthdays and Anniversaries — present but visually quiet.
  5. Bottom tab bar preserved — the existing nav pattern users already know stays intact.

Redesign mockup built in HTML/CSS · not a real product screenshot

Execution · 05 — Mobile Salary Screen

The donut chart problem, amplified on a phone.

On a phone, the chart takes half the viewport to deliver less information than the list below it. The fix is simpler than it looks: remove the chart, promote the list, add a total.

⚠ Problem — Actual Mobile Salary, Annotated

Half the screen says nothing. The answer is below the fold.

⚠ Annotated Problem
Mobile salary annotated — donut chart failure
1
2
3
4
5
  1. “Salary Summary” header — no month mentioned, no greeting, no net-pay headline to orient the user.
  2. Donut chart takes 50% of the viewport to show less information than the list below. Pure visual noise on a phone.
  3. 9.4% and 12.5% labels physically collide on the bottom segment — unreadable at phone resolution.
  4. Amount values right-aligned with no total line — no number on screen answers “what did I earn this month?”
  5. Line-item list is correct — the fix is simply to promote this list and kill the chart above it.
✓ Proposed Fix — Net Pay Card + Linear List

One number at the top. Every line readable on its own.

The chart is gone. The net pay is the first thing users see. Every earnings line stays, with a short context hint underneath so users who aren't payroll-literate understand what each component is.

11:2844% ▣
Your Salary Breakdown
July 2023 · paid 1 Aug
3
Net pay this month
₹ 32,000
Credited · A/C ending 4471
Summary
Earnings
Deductions
Earnings · 4 items
Basic PayFixed core salary
₹ 20,000
House Rent AllowanceHRA
₹ 4,000
Dearness AllowanceDA
₹ 3,000
Special AllowanceVariable
₹ 5,000
Dashboard
Salary
Tax
Inbox
Menu
  1. Title says “Your Salary Breakdown” with month + pay date — full orientation in one line.
  2. Net pay card with ₹32,000 and credit account — the literal answer to the user's question, first and largest.
  3. Tabs preserved — Summary / Earnings / Deductions structure stays familiar.
  4. Each line has a context hint — “HRA · tax-exempt up to limit”, “DA · cost-of-living adjustment” — no jargon mystery.
  5. No chart anywhere — the data is the chart. The numbers are the visualisation.

Redesign mockup built in HTML/CSS · not a real product screenshot

07
Microcopy Transformation
Before vs After — word by word

Microcopy

Every word replaced had a specific reason.

Microcopy isn't about tone. It's about removing the cognitive work users shouldn't have to carry.

CTA Labels
Before
Submit
Generic. What is being submitted? What happens next? Users hesitate because they can't predict the outcome.
After
Save your details
Verb + object + implied safety. “Save” signals reversibility. “Your” reduces distance. Outcome is clear before tapping.
Before
Proceed
Proceed to what? Entirely context-free. Users must remember their own position in the flow.
After
Continue to salary summary
Destination named. Users know exactly what the next screen will show. No navigation anxiety or second-guessing.
Before
Details
Details of what? A noun without a verb. Users don't know if this is a link, button, or label.
After
View full breakdown
Action verb + specific outcome. “Breakdown” signals structure. No ambiguity about what tapping does.
Section & Screen Labels
Before
Earnings Pie Chart
Names the format. Users don't care how data is visualised — they care what it means about their pay.
After
Your Salary Breakdown
Names the answer, not the container. “Your” personalises it. “Breakdown” promises structure and clarity — not a chart.
Before
Remark if any
What kind of remark? About what? For whom? Users have no idea what to write, or if they even should.
After
Add attendance note (optional)
“Attendance note” defines the context. “(optional)” removes pressure. Users know exactly what this field is for.
Before
Hide all seen events
What events? On which screen? “Hide” sounds like something disappears. Users avoid unclear system actions.
After
Mark all notifications as read
Describes the actual action precisely. “Notifications” is the familiar mental model. No learning curve.
Status & Confirmation Messages
Before
[No confirmation after Clock In]
After tapping Clock In, no acknowledgment. Did it register? Should I tap again? Silence creates re-taps and anxiety.
After
You're clocked in. Have a great day, Rohan.
Confirms the action, uses the user's name, adds warmth. Eliminates the “did it register?” doubt completely.
Before
[No orientation on salary screen load]
Users landed on a dense data screen with zero orientation — no greeting, no summary, no sense of context.
After
Here's your summary for July 2023.
One sentence. Costs nothing in space. Removes the “where am I?” cognitive tax on every page load.
08
System Thinking
Tool → Assistant

System Framing

The work wasn't about individual screens. It was about what the system says when users arrive.

Four shifts in how the product communicates — not what it shows, but how it speaks.

Before
Tool
After
Assistant
The app processed tasks silently. Now it guides users through them — explaining what's happening and what comes next before they have to ask.
Before
Data
After
Meaning
Numbers existed on screen. Now they are interpreted — grouped, labelled, contextualised so users understand without needing to calculate.
Before
Interface
After
Guidance
Screens presented options at equal weight. Now they sequence users through priorities — what matters first, what to do next, and why.
Before
Silence
After
Confirmation
Actions completed without acknowledgment. Now every meaningful action gets a response — building trust one interaction at a time.
09
Impact
Honest outcomes

Impact

No fabricated metrics. Real qualitative outcomes.

This was a UX writing and critique engagement. Impact is measured in clarity — not dashboard numbers.

01
Reduced Salary Ambiguity
The #1 user complaint addressed by removing the chart, surfacing the ₹32,000 net pay, and creating a readable linear breakdown. Users get the answer without arithmetic.
02
Clearer Action Paths
Every CTA now tells users what happens next — not just that something will happen. Reduced hesitation, re-reading, and flow abandonment.
03
Safer Navigation
Menu restructure separated destructive system actions from daily navigation. Users can explore the app without fear of accidentally clearing their attendance data.
Scope note: This case study documents UX writing improvements and proposed UX enhancements completed during a two-month engagement. Not all changes were implemented in the live product. The annotated screens show problems identified on the actual product; the redesign mockups show the solutions proposed. UX critique and writing proposals are contributions regardless of implementation status — the thinking is the work.
10
Key Takeaway
What this project taught

10 — Key Takeaway

UX writing is not about making interfaces sound better. It is about making systems understandable.

HR365 had all the features. The salary data was accurate. Attendance tracking worked. The app didn't fail because it was broken — it failed because it assumed users could interpret raw output without guidance. The intervention wasn't cosmetic. It was structural: every label, CTA, section title, and status message is a decision about how much cognitive work users have to carry. This project was about putting that weight back on the system, where it belongs.

A note on working with imperfect products:
Negative reviews reflect the product experience — not the UX writer's contribution. Real UX work happens inside products with real constraints, partial buy-in, and incomplete implementation. The value of this case study is not the app's rating. It is the thinking that went into diagnosing what was wrong and proposing what could be better. That thinking transfers to any product, any team, any stage.