Seeking Entry-Level Data / Business Analyst Opportunities

Anushka
Sharma

Data & Business Analyst · JIIT, CGPA 9.3

Turning Data into
Product Decisions.

0
Usability Improvement
↑ 68 → 94 score (Gametosa)
0
Records Cleaned & Analyzed
↑ Ops team reporting
0
Collaborated With
↑ Product, Data & Engineering
9.3
Academic Excellence
↑ JIIT, CS Honours
Usability Score — Gametosa
94 +38%
Live Reporting — Zidio
500K records
About

Data Analyst. Product Storyteller.
Clear Communicator.

I'm a B.Sc. (Hons.) Computer Science student at JIIT (CGPA 9.3) focused on data and business analysis. I analyse product data to identify patterns, quantify problems, and recommend measurable improvements — then present findings and recommendations to product and operations teams.

My work spans retention analysis, KPI dashboards, cohort reporting, and product usability research. I've cleaned and analysed large datasets, built BI dashboards for the operations team, and worked with product and engineering teams to clarify requirements and frame measurable questions. Comfortable working with ambiguous product problems and translating them into measurable hypotheses.

SQL Power BI Cohort Analysis KPI Reporting Retention Analysis Hypothesis Testing Churn Analysis Cross-team Communication
9.3
CGPA — Jaypee Institute of Information Technology
500K
Records cleaned and structured for operational reporting
2
Industry internships with quantified business impact
+38%
Usability improvement delivered at Gametosa eSports

Work Driven by Metrics,
Not Just Deliverables.

Impact first.
Method second.
📊
Customer Retention & Churn Analysis
SQL · Retention Cohorts · Segmentation
Month 3
Cohort drop-off peak identified in analysis
⚠ Retention Risk
3 segs
High-risk user segments surfaced via SQL
→ Actionable
High-risk segments
Methodology
SQL Segmentation Cohort Analysis Retention Patterns User Segmentation
Key Finding Analyzed customer lifecycle data to identify retention patterns. Surfaced 3 distinct user segments with diverging engagement trends and recommended targeted re-engagement strategies for the highest drop-off cohort.
View project on GitHub View dashboard report
🎯
Product Usability Optimization
Gametosa eSports · UX Analysis · API Validation
94
Usability score post-fix (from 68)
↑ +38% improvement
Improved
Page response time on product detail pages
↑ Flagged to eng team
Score: 68 Score: 94 Fix applied
Methodology
WCAG 2.1 Guidelines Drop-off Analysis API Benchmarking React Optimization
Business Impact 26-point usability improvement (+38%) improved the usability and accessibility of product pages. Accessibility improvements based on WCAG 2.1 guidelines — covering contrast ratios, keyboard navigation, and semantic HTML — brought pages closer to full compliance.
📈
Operational KPI Dashboard & Data Reporting
Zidio Development · Power BI · Ops Reporting
Real-time
Power BI dashboards deployed for the ops team
↑ Decision visibility
500K
Records cleaned, structured & analysed
↑ Data quality
KPIs live → Data volume over time
Methodology
Power BI KPI Dashboarding Data Cleaning Ops Reporting
Business Impact Built real-time Power BI dashboards giving the operations and product team instant visibility into KPIs. Cleaned and structured 500K+ records to produce reliable, analysis-ready data — reducing time spent on manual reporting and making performance metrics easier to monitor.
Skills

Tools of the Trade

Data Analysis
SQL
Excel & Google Sheets
Data Cleaning & Structuring
Exploratory Data Analysis
BI & Visualisation
Power BI
Dashboard Design
KPI Reporting
Google Analytics
Product Analytics
Funnel Analysis
Retention & Cohort Analysis
Usability Metrics
User Journey Mapping
Programming & Comms
Python (Pandas, NumPy)
Basic Statistics
Data Storytelling
Cross-team Communication
Approach

How I Think Through
Every Problem

A structured, repeatable mental model for turning business questions into data-backed recommendations.

01
Define the Metric
What does success look like? Align on the north star metric before opening any dataset.
02
Analyze the Data
SQL queries, cohort cuts, distribution checks. Find patterns and anomalies without bias.
03
Identify Root Cause
Drill from symptom to cause. Segment by channel, user type, time window until signal is clear.
04
Recommend Action
Prioritized, effort-mapped recommendations. Trade-offs clearly articulated for the product and business team.
05
Measure Impact
Define leading and lagging indicators. Track post-launch. Close the feedback loop.
Experience

Where I've Built
Real-World Impact

Jul 2025 — Present
Software Development Intern (Product Analytics Focus)
GAMETOSA eSports · Remote
Problem
Gaming e-commerce platform had a usability score of 68, with high drop-off rates and WCAG compliance gaps limiting accessibility for a broad audience.
Action
Contributed to UI/UX improvements, conducted drop-off funnel analysis, improved accessibility compliance based on WCAG 2.1 guidelines, and collaborated with backend engineers to validate API response SLAs.
Result
Usability score improved 68 → 94 (+38%). Page response time on product detail pages improved following engineering fixes. Brought pages closer to WCAG accessibility compliance.
May — Aug 2024
Data Science Intern — Analytics & Reporting
Zidio Development · Remote
Problem
The operations team had no real-time visibility into performance. Data was unstructured across 500K+ records with no consistent reporting layer in place.
Action
Deployed interactive Power BI dashboards for the operations and product team. Cleaned, structured, and analysed 500K+ audio records. Assisted in preparing datasets for use in an internal ML project.
Result
The product and ops team gained real-time KPI dashboards enabling faster decisions. Data quality improved significantly, giving the team reliable, analysis-ready datasets for ongoing reporting.
Case Study

Customer Churn & Revenue Risk Analysis
— RevenueGuard

A full walkthrough of how I identified churn-prone customer segments, quantified revenue exposure, and translated findings into actionable retention recommendations.

01 · Problem
02 · Data Used
03 · Metric Defined
04 · Analysis Steps
05 · Insights Found
06 · Recommendation
07 · Outcome
Business Problem

A subscription dataset had no visibility into which users were at risk of leaving — or how much revenue was exposed.

A subscription-style customer dataset lacked visibility into which users were at risk of leaving and how much revenue was exposed to churn. The objective was to identify churn-prone customer segments and quantify potential revenue loss — turning raw customer data into a prioritised retention view.

Business Question "Which customers are most likely to churn, and what revenue is at risk if they do?"
Data Used

Structured customer dataset covering behaviour, contract, and billing attributes.

The analysis used a structured customer dataset containing key attributes across tenure, billing, and support behaviour. Each field was selected for its relevance to churn likelihood — contract type and tenure as structural risk indicators, monthly charges and payment method for revenue sizing, and support usage as a signal of friction or dissatisfaction.

📅
Customer tenure
💳
Monthly charges & payment method
📄
Contract type & support usage
🔴
Churn status (target variable)
Metric Defined

Primary focus: churn probability by segment. Secondary: revenue at risk.

Before any analysis, the metrics were defined clearly so findings would map directly to retention decisions. The primary metric was customer churn probability — which customers are most likely to leave based on their profile. Supporting metrics gave the business dimension: how much revenue was concentrated in high-risk segments, and whether any high-value customers were disproportionately exposed.

Primary
Churn Probability
Segment-wise churn rate
Supporting
Revenue at Risk
High-value customer loss exposure
Analysis Steps

Five structured steps from raw data to retention-ready insight.

The analysis followed a structured sequence to avoid jumping to conclusions — cleaning first, then understanding distributions, then segmenting and sizing the risk.

1
Data cleaning & validation: Cleaned and validated the dataset using Python (Pandas) — handled missing values, standardised data types, and checked for inconsistencies in the churn status field.
2
Exploratory data analysis: Examined distributions and anomalies across key fields — tenure spread, charge ranges, contract type breakdown — to understand the shape of the data before forming hypotheses.
3
Customer segmentation: Applied SQL-style filtering logic to group customers by contract type, tenure band, and payment method. Compared churn rates across each segment to identify which profiles were most at risk.
4
Retention cohort building: Built retention cohorts based on tenure and contract type to surface which early-stage or flexible-contract customers were exiting most quickly.
5
Revenue at risk calculation: Combined churn likelihood with each customer's monthly charge to estimate the revenue exposure concentrated in high-risk segments — sizing the problem in business terms.
Insights Found

Four clear patterns — each pointing to a different retention lever.

The segmentation surfaced four distinct findings, each with a different retention implication. Rather than a single root cause, the data showed that churn risk was concentrated across specific contract and tenure profiles — meaning retention interventions would need to be targeted, not blanket.

1
Month-to-month contract users showed the highest churn rate — no long-term commitment meant the lowest switching cost.
2
Short-tenure users (under 6 months) formed the most unstable cohort — the early period was the highest-risk window for dropout.
3
Auto-pay customers had significantly lower churn — removing friction from billing correlated with stronger retention.
4
A small group of high-value users contributed disproportionate revenue risk — their churn exposure was outsized relative to their share of the customer base.
Key Insight Churn was not evenly distributed. Targeting the right segments — early-tenure and month-to-month users — would address the majority of risk with focused effort.
Recommendation

Three targeted retention actions, each matched to a specific at-risk segment.

Recommendations were grounded directly in the segment findings — each action mapped to a specific at-risk profile rather than a generic retention campaign.

R1 · Early-Tenure Users
Introduce structured onboarding engagement for the first 3 months — the window where dropout risk is highest. Proactive check-ins or usage prompts during this period could reduce early exits.
R2 · Month-to-Month Contract Users
Offer incentives for annual contract conversion — reducing churn risk by increasing switching cost. Even a small shift from flexible to fixed contracts in this segment would meaningfully reduce revenue exposure.
R3 · High-Value At-Risk Users
Implement targeted outreach to the small cohort of high-value customers with elevated churn signals. Personalised communication or account review could protect a disproportionately large share of revenue.
Measured Outcome

Interactive dashboard produced — churn segments and revenue exposure made visible in one view.

Produced an interactive Power BI dashboard summarising churn segments and revenue exposure, allowing business teams to quickly identify risk areas and prioritise retention strategies. The dashboard translated the full analysis into a format decision-makers could act on directly — without needing to interpret the underlying data themselves.

4
At-risk segments identified and sized
3
Targeted retention actions recommended
1
Interactive dashboard delivered
What I Delivered
🐍
Python Analysis (Pandas)
Data cleaning, EDA, and segmentation logic — full code available in the GitHub repository.
📊
Power BI Dashboard
Interactive report visualising churn segments and revenue at risk — viewable in the dashboard PDF linked below.
📐
Problem → Metric → Insight Map
Structured document linking each business question to a specific metric, analysis step, and retention recommendation.