Namit Singh

Data & Business Analyst

Turning tangled data into clear decisions.

I help teams see what the numbers are really saying, translating dashboards, models and messy spreadsheets into decisions people can act on.

3 yrs Analytics SQL · Python · BI · Automation · Claude
Download CV ↓
Scroll
Open to new opportunities
About

A partner to the business, not just a report factory.

I sit between the data and the people who need it, asking the sharper question before writing the query, then delivering answers that change what the team does next.

My work spans the full analytics loop: scoping the real business question, wrangling and modeling the data, building dashboards and forecasts, and most importantly communicating the "so what" to stakeholders who don't live in spreadsheets.

I care about clean data, honest charts, and recommendations that hold up in the meeting after the meeting.

FocusProduct · Growth · Ops analytics
ApproachQuestion-first, decision-driven
Skills & tools

The stack behind the insights.

01

Data engineering

SQLPythonPandasNumPyETLAzure
02

BI & visualisation

Power BITableauAI in Excel (Power Query, PivotTables, advanced functions, lookups)Dashboard designKPI reporting
03

Business analysis

Requirements elicitationStakeholder managementProcess mappingUser storiesMoSCoW prioritisationWorkshop facilitation
04

Analytics

StatisticsForecastingA/B testingCohort analysisRoot cause analysisScenario analysis
05

Machine learning

scikit-learnLogistic RegressionRandom ForestXGBoostK-Means ClusteringPCA
06

Delivery & change

PRINCE2®AgileScrumJIRAKanbanRoadmapping
07

Communication

Data storytellingExecutive reportingDocumentationPresentationsTrainingFacilitation
Experience

Where the numbers met the work.

Nov 2026 – Present
Business Analyst
Keele University
  • Joined Keele's Transformation team to help redesign how professional services work for students and staff, currently focused on the university's ServiceNow implementation.
  • Mapping how students and staff use services, online and in person, through journey maps, personas and user stories.
  • Assessing current processes and helping define what the future service should look like.
  • Turning user needs into clear requirements, acceptance criteria and specifications.
  • Using data to show where services fall short and what to fix first.
  • Working with teams across the university so changes fit how people really work.
Sep 2025 – Sep 2026
Business Analyst
Primark
  • Elicited business, process and reporting requirements from HR stakeholders, Store Managers and People & Culture Business Partners, clarifying what decisions each request supported rather than building what was first asked for.
  • Analysed the existing reporting process end to end and replaced manual spreadsheet preparation with standardised reporting, reducing recurring reporting time by 35%.
  • Built and maintained validation and audit controls across 550+ employee records, covering required-field completion, format consistency and record currency, maintaining 96% data compliance and investigating exceptions at source.
  • Supported delivery through the business-wide rollout, coordinating requirements, reporting and data-quality activity across the programme.
  • Delivered training to 20 line managers, supporting business readiness, user adoption and improved understanding of data and process requirements.
  • Produced reporting packs on headcount, absence and turnover with written commentary translating workforce data into recommendations for stakeholders.
Sep 2024 – Aug 2025
Business Analyst
Pratapvasini — Digital transformation & migration programme
  • Facilitated stakeholder workshops and one-to-one interviews across multiple workstreams to document the As-Is state, including process steps, decision points, information flows, dependencies and pain points.
  • Analysed a fragmented legacy information landscape spanning physical registers, paper forms, identification documents, invoices, operational, logistics and staff records, defining the To-Be digital process and data requirements.
  • Produced source-to-target data mappings covering business definitions, transformation rules, mandatory and optional fields, validation rules and data ownership, working with Data SMEs to ensure technical mappings reflected business meaning.
  • Established data quality rules across completeness, accuracy, consistency, uniqueness and validity, including agreed approaches for identifying duplicates and handling incomplete records.
  • Determined source-of-truth and data ownership where information existed across conflicting sources, documenting data lineage and downstream reporting dependencies.
  • Translated business needs into user stories, acceptance criteria and functional documentation, maintaining requirements traceability from business need through delivery, testing and UAT.
  • Prepared UAT test scenarios covering valid, exception and migrated-record cases, logged defects, coordinated resolution with Data SMEs and retested to closure.
  • Reconciled migrated record populations against source data, investigating and explaining variances rather than simply matching totals.
  • Managed Agile delivery in JIRA, maintaining sprint boards, backlog items, dependencies and progress reporting across parallel workstreams.
Dec 2023 – Jul 2025
Retail Operations Supervisor
Poundland — Part-time alongside postgraduate study and analyst work
  • Supervised day-to-day store operations, coordinating colleagues, task allocation and workflow in a fast-paced retail environment.
  • Supported the rollout of the Poundland Rewards app, coordinating store readiness, colleague communication and customer adoption during launch.
  • Provided operational support to management teams across nearby stores, investigating issues, implementing operational changes and sharing best practice.
  • Analysed store performance, customer activity and stock availability to identify issues and support management decision-making.
  • Managed staff scheduling and resource allocation against trading patterns and peak periods, balancing customer demand with workforce capacity.
Jun 2022 – Aug 2023
Jr. Business Analyst
MyCaptain — EdTech · Multi-project Agile delivery
  • Facilitated stakeholder discussions and workshops to clarify project and reporting requirements and align business, project and technical stakeholders.
  • Oversaw sprint tracking, backlog items, timelines and dependencies across multiple Agile projects in JIRA, providing structured delivery visibility.
  • Built and maintained programme KPI dashboards covering delivery progress, backlog, milestones and project health to support management decision-making.
  • Validated delivery data and identified inconsistencies in how teams applied status definitions, establishing agreed definitions to improve cross-project reporting and comparison.
  • Translated stakeholder questions into reporting requirements and designed separate executive and project-level views, ensuring the same underlying data supported different decision-making needs.
Forage job simulations

Practicing the actual work.

Forage · Data Analytics

Data Analytics Job Simulation

Quantium
  • Cleaned and prepared a 264,000+ row transaction dataset in Python/R, converting date formats, removing non-category (salsa) products and bulk-buy outliers, and engineering pack-size and brand features from raw product names, then merged it with 72,000+ customer records to build an analysis-ready dataset.
  • Segmented $1.8M in annual chip sales across 21 lifestage × affluence customer groups, identifying Mainstream Young Singles/Couples as the highest-opportunity segment (largest by customer count and least price-sensitive) and translating the findings into targeted brand, pack-size and pricing recommendations.
  • Built a reusable control-store selection function benchmarking 260+ stores against trial stores using Pearson correlation and a scaled magnitude-distance metric, ensuring statistically valid, like-for-like comparisons for uplift testing.
  • Measured trial-layout impact with t-tests on scaled control baselines, proving a statistically significant +12% to +26% sales uplift across three trial stores and diagnosing the driver as increased customer acquisition rather than higher spend per customer.
  • Consolidated insights into a client-ready report for the Category Manager using the Pyramid Principle, pairing consistent data visualisations with clear callouts and actionable recommendations to inform the half-year category strategy.
View certificate ↗
Forage · Quantitative Research

Quantitative Research Job Simulation

J.P. Morgan
  • Built a natural gas price estimation model in Python (pandas, NumPy) that decomposed 4 years of historical price data into trend and seasonal components, enabling on-demand price lookups for any date and 12-month forward extrapolation; validated by exactly reproducing all 48 historical data points.
  • Designed a storage contract pricing engine that computes the fair value of arbitrary gas injection/withdrawal schedules by netting commodity, storage, transport, and handling cash flows while enforcing facility rate and capacity constraints; validated against a hand-calculated reference case and automated constraint-violation tests.
  • Trained and compared logistic regression, gradient boosting, and random forest models to predict probability of default (PD) on a 10,000-loan portfolio, selecting logistic regression for its interpretability and audit-readiness; cross-validated performance (AUC ≈ 0.9998) and reconciled modeled expected loss against realized portfolio losses to within 0.01%.
  • Engineered an exact dynamic programming algorithm (O(n²r) partitioning) to convert continuous FICO scores into optimized categorical rating buckets for a downstream PD model, maximizing a log-likelihood objective over default outcomes and confirming global optimality against brute-force search; diagnosed and corrected a bucket-count edge case that produced non-monotonic risk ratings.
View certificate ↗
Forage · Digital Transformation

Digital Transformation Job Simulation

BCG
  • Benchmarked best-in-class F&B mobile apps (Luckin Coffee, Domino's) and built a client discussion deck mapping every app feature to a customer-journey stage and to the same-store-sales driver it moved: purchase frequency, ticket size, or first-party data capture.
  • Analysed a 3,300-transaction dataset spanning a baseline week and two personalisation pilots, building a live Excel model computing revenue, gross profit, average ticket and profit per transaction, with store-count and traffic-persistence sensitivities for national scale-up.
  • Found the higher-revenue pilot (+30% revenue) was volume-driven and margin-dilutive, with profit per transaction falling 0.5%, while the lower-revenue pilot delivered a structural +6.9% ticket size and +4.8% profit per transaction; recommended scaling the second.
  • Identified that neither pilot ran a concurrent control group, meaning reported gains reflected week-on-week variation rather than measured incremental effect; specified the customer ID, holdout design and store-level tagging required before results could support a business case.
  • Generated and MECE-categorised personalisation options using a WHO × WHAT × WHEN message framework, then scored five candidate features against customer experience, revenue impact, implementation cost and scalability to produce a weighted priority ranking.
  • Translated the recommendation into a two-week agile sprint plan in Trello: 3 epics, 10 user stories and 12 tasks sequenced across data science, engineering, UX and legal workstreams, with measurement infrastructure deliberately built ahead of feature work.
View certificate ↗
Forage · Data Visualisation

Data Visualisation: Empowering Business with Effective Insights

Tata Consultancy Services
  • Completed a TCS data visualization simulation as an external analytics consultant, advising a retail company's CEO and CMO on revenue performance and market-expansion strategy.
  • Cleaned and validated a 540,000+ row transaction dataset by defining data-quality rules (removing invalid quantities and pricing errors) to ensure analysis-ready, audit-traceable data for executive reporting.
  • Drafted 8 stakeholder-discovery questions differentiating CEO priorities (revenue growth, profitability, expansion risk) from CMO priorities (customer segmentation, retention, campaign timing), translating open-ended business goals into concrete analytics requirements.
  • Designed and built 4 executive-facing visualizations (time-series trend, dual-axis comparison, ranked bar chart, treemap) to surface seasonal revenue patterns, top-performing international markets, and customer revenue concentration.
  • Delivered a 5-minute stakeholder presentation translating findings into a data-backed expansion recommendation, identifying five priority markets where revenue and demand signals aligned.
View certificate ↗
Forage · Data Analytics

Data Analytics Job Simulation

Deloitte Australia
  • Completed a Deloitte forensic technology and data analytics job simulation, analysing IoT telemetry and compensation data to surface operational and HR risks for a client, Daikibo Industrials.
  • Built an interactive Tableau dashboard from a 160,000-record JSON telemetry feed, engineering a calculated "downtime" measure and a linked factory-to-device filter that pinpointed Factory Seiko's Laser Welder as the single largest source of machine downtime (480 minutes/month).
  • Classified 37 role-level pay-equality scores in Excel into a three-tier fairness scale, exposing that ~49% of roles fell outside the fair range and flagging 7 "highly discriminative" cases for leadership review.
View certificate ↗
Forage · Integrated Consulting Group

Integrated Consulting Group Job Simulation

Oliver Wyman
  • Diagnosed underperformance across a $100M insurance book, tracing an 8-point combined ratio deterioration (97% to 105%) to a single line: Auto, at 110% on 40% of premium, against 97.8% for the remaining portfolio.
  • Separated claims cost from overhead as the true driver, Auto's 82% loss ratio ran 12–23 points above every other line while its expense ratio stayed below the divisional average, redirecting the recommendation away from cost-cutting toward pricing and risk selection.
  • Showed the division's 2-point expense-ratio rise was more than accounted for by a 56% increase in IT spend, worth 2.3 points of premium, reframing it as one-off migration cost rather than structural inefficiency.
  • Identified a 2.3-point reconciliation gap between product-line and divisional combined ratios and flagged unsourced benchmark figures in the team deck, keeping unverified numbers out of the client-facing version.
  • Designed a four-tile executive dashboard from nine candidate KPIs, sequenced as a diagnostic chain, and authored the client recommendation slide proposing a 90-day Auto pilot against a matched control in place of a book-wide rollout.
View certificate ↗
Education & certification

Grounded in rigor.

Education
MSc Business Analytics
University of Warwick
2023 – 2024
B.E. Electronics Engineering
University of Mumbai
2019 – 2023
Certifications
PeopleCert
PRINCE2 7 Foundation
PeopleCert · Issued Aug 2025
↗
Featured work

Recent projects.

View repo ↗
AI Opportunity Discovery & Prioritisation — n8n workflow overview
n8n workflow: data ingest, AI processing, six-criteria scoring, and routing
Responsible-AI governance routing: risk-based decision gateway
Full n8n editor canvas of the opportunity prioritisation workflow
AI automation · BA · GitHub

AI opportunity discovery & prioritisation

An n8n workflow that turns messy AI-idea intake into governed decisions: a local LLM produces a structured BA assessment, a weighted six-criteria model scores each idea, and a switch routes it — with governance checked before value, so high-risk ideas never skip review.

Stack: n8n · Docker · Ollama (LLM) · JavaScript · Responsible-AI governance
View repo ↗
View repo ↗
Medallion architecture
Bronze
Silver
Gold
CRM + ERP → SQL Server
Data warehouse architecture diagram
Data flow diagram
Star schema data model
Data integration diagram
Data engineering · GitHub

Modern data warehouse from scratch

Built a modern data warehouse using SQL Server, covering the full ETL pipeline, dimensional data modelling, and analytics-ready layers.

Stack: SQL Server · ETL · Dimensional modelling
View repo ↗
View repo ↗
RFM segmentation model
R
F
M
RecencyFrequencyMonetary
RFM · retention · basket affinity
SELECT category_a, category_b, COUNT(*) AS co_purchases
FROM basket_pairs GROUP BY category_a, category_b;
SQL analytics · GitHub

Sales warehouse, gold-layer analytics

Advanced SQL analytics on top of a data warehouse's gold layer — exploration, ranking, time-series, segmentation and RFM, built with window functions, CTEs and T-SQL reporting views.

Stack: T-SQL · Window functions · RFM segmentation
View repo ↗
View repo ↗
3,900-row pipeline
CSV
pandas
SQL Server
Tableau
$233K
Total revenue
3,900
Customers
3.75
Avg. rating
Consumer shopping trends overview dashboard
Revenue and spending dashboard
Products and categories dashboard
Segmentation and loyalty dashboard
Behavioral and cross-factor dashboard
Retail analytics · GitHub

Customer shopping trend analysis

End-to-end retail analytics: cleaned a 3,900-row consumer shopping dataset in Python/pandas, then ran SQL Server EDA covering revenue drivers, customer segmentation and product performance.

Stack: Python · pandas · SQL Server · EDA
View repo ↗
View repo ↗
Sarcasm-corrected sentiment
"Local man reads one article, now an economist."Reads positive
😏 Sarcasm detectedCorrected → negative
96.7%
Accuracy (Bi-LSTM)
0.964
F1 score
4
Architectures compared
NLP · GitHub

Sentiment analysis with sarcasm detection

A deep-learning sarcasm detector on news headlines, combined with VADER sentiment to correct misleading labels — comparing LSTM, Bi-LSTM, Attention-LSTM and Transformer models.

Stack: Python · LSTM / Transformers · VADER
View repo ↗
View repo ↗
Issue & return workflow
Shelf
Member
Branches · employees · CTAS reports
Library management system entity-relationship diagram
SQL · GitHub

Library management system

A relational schema for branches, employees, members, books, and issue/return tracking — full CRUD, CTAS summary tables, and stored procedures for issuing/returning books and branch performance reporting.

Stack: MySQL · CTAS · Stored procedures
View repo ↗
View repo ↗
Automated ETL · GitHub Actions
7,684
Products sold
70/30
Prime split
6h
Run schedule
E-commerce entity-relationship diagram
SQLite ingestion query
Average spend by membership status chart
Top states by customer count chart
Data engineering · GitHub

End-to-end e-commerce data pipeline

A simulated university e-commerce platform: 3NF schema design, synthetic data generation, automated GitHub Actions ETL running every 6 hours, and R/Quarto analysis of purchasing behaviour.

Stack: R · SQLite · GitHub Actions · Quarto
View repo ↗
View repo ↗
Retrieval-augmented generation
MiniLM
FAISS
GPT-2
“What is the significance of Thane station?”
NLP / RAG · GitHub

RAG system for Mumbai local travel

A retrieval-augmented generation pipeline answering natural-language questions about Mumbai's railway stations — MiniLM embeddings, FAISS nearest-neighbour search, and GPT-2 grounded generation.

Stack: Sentence-Transformers · FAISS · GPT-2
View repo ↗
View repo ↗
11 classifiers benchmarked
SVC
Log. Reg
Naive Bayes
XGBoost
Selected for precision on TF-IDF
NLP / ML · GitHub

Email & SMS spam detection

A full NLP pipeline — cleaning, TF-IDF vectorization, and a benchmark of eleven classifiers plus Voting/Stacking ensembles. Multinomial Naive Bayes was selected and serialized for its precision.

Stack: scikit-learn · XGBoost · NLTK · TF-IDF
View repo ↗
View repo ↗
Sales by daypart
Morning
Afternoon
Evening
CASE-based shift segmentation
SQL · GitHub

Retail sales analysis

A SQL project: database setup, null-handling data cleaning, and business queries covering best-selling months, top 5 customers, and morning/afternoon/evening sales shifts.

Stack: SQL · Window functions · Data cleaning
View repo ↗
View repo ↗
Content-type split
Movies
TV Shows
Top countries · genres · keywords
SQL · GitHub

Netflix catalogue analysis

15 SQL queries exploring Netflix's movie/TV catalogue — content-type split, top countries and genres, longest titles, and keyword-based content categorisation using CTEs and recursive queries.

Stack: SQL · CTEs · Recursive queries
View repo ↗
Tableau Public

Interactive dashboards, live on Tableau

Explore the visualizations behind these projects — filterable, published dashboards you can click through yourself.

View on Tableau Public ↗
Writing

Notes from LinkedIn.

Let's connect

Have a question your data can't answer yet?

↗ Email ↗ LinkedIn /in/namitt ↗ GitHub @Namitt ↗ Tableau Public profile ↓ Download CV PDF resume

Send a message