MBA Analytics · Available Now

LESLEY
SHIRI

Business Analytics · Finance Intelligence · Decision Support

MBA · MCom · BCom · National Diploma. Four qualifications, six professional roles, and 8+ years turning financial and operational complexity into reporting, controls, and decisions leaders can act on.

MBA — Clarkson University MCom · BCom · N.Dip Payroll Specialist Business Analyst Finance Analyst Revenue Analyst Operations Specialist Accountant
View Strategic Portfolio Review Executive Resume Schedule Hiring Conversation
Scroll to explore
MBA Analytics✦MCom✦BCom Accounting✦National Diploma✦Payroll Specialist✦Business Analyst✦Finance Analyst✦Revenue Analyst✦Operations Specialist✦Accountant✦Python✦Power BI✦Tableau✦SQL✦Oracle Cloud✦ MBA Analytics✦MCom✦BCom Accounting✦National Diploma✦Payroll Specialist✦Business Analyst✦Finance Analyst✦Revenue Analyst✦Operations Specialist✦Accountant✦Python✦Power BI✦Tableau✦SQL✦Oracle Cloud✦
01 — Identity

Where financial precision
meets data intelligence

I'm a Business and Finance Analyst with 8+ years of experience across payroll, accounting, finance analysis, revenue analytics, operations, and business analysis. My work turns dense business data into reporting, controls, and recommendations that improve decision quality.

My academic foundation combines a National Diploma in Accounting (Masvingo Polytechnic, Zimbabwe), a BCom Honours in Accounting, a Master of Commerce (Midlands State University, Zimbabwe), and an MBA in Business Analytics at Clarkson University, New York.

I bring together accounting rigour, operational understanding, and applied analytics, from healthcare prediction models to payroll audit workflows and executive revenue reporting. Four qualifications, six roles, one clear value proposition.

Education
MBA
MBA — Business Analytics
Clarkson University, New York · In Progress
MCom
Master of Commerce — Accounting
Midlands State University, Zimbabwe
BCom
BCom Honours — Accounting
Midlands State University, Zimbabwe
N.Dip
National Diploma — Accounting
Masvingo Polytechnic, Zimbabwe
Career Timeline
2022–2024
Revenue Analyst
May 2022 to Dec 2024
2019–2022
Payroll Accountant
Jan 2019 to May 2022
2014–2019
Senior Financial Analyst
Jul 2014 to Jan 2019
2012–2014
Accounts Clerk
Jan 2012 to Jul 2014
Identity
Lesley Shiri portrait

Lesley L. Shiri

Financial precision with analytical imagination. Built for insight, execution, and measurable impact.

MBA AnalyticsFinance + Data6 Career Roles
4
Qualifications
6
Career Roles
8+
Yrs Finance
100K
Patient Records
0.72
Best F1 Score
$130K
Payroll Exposure Flagged
02 — Credentials

Built on four
academic foundations

MBA
Apex Degree
Master of Business Administration
MBA · Business Analytics · Clarkson University, NY

Business strategy, data-driven decision-making, ML pipelines, and analytics leadership — the capstone unifying every prior qualification.

In Progress
MCOM
Postgraduate
Master of Commerce
MCom · Accounting · Midlands State University, ZW

Advanced accounting theory, financial research methodology, and commerce economics — the depth to challenge assumptions.

Step 3 of 4
BCOM
Undergraduate
BCom Honours — Accounting
BCom · Accounting · Midlands State University, ZW

Financial accounting, auditing, taxation, management accounting — the complete commercial framework of modern business.

Step 2 of 4
N.D.
Foundation
National Diploma
N.Dip · Accounting · Masvingo Polytechnic, ZW

Financial statements, cost accounting, and business law — where the journey began and the discipline was forged.

Step 1 of 4
Academic Progression
N.Dip→BCom (Hons)→MCom→MBA Business Analytics
Four qualifications. One vision.
03 — Capabilities

Skills that
deliver results

Core Language
Python & Machine Learning
Pandas / NumPy95%
Scikit-Learn90%
Random Forest / Gradient Boosting88%
Neural Networks82%
Credentials
4+6
Four qualifications · Six roles · The rare combination of accounting rigour and data science firepower.
Finance & Accounting
Financial Intelligence
Oracle Cloud HCMPayroll ProcessingAccounts PayableAdvanced ExcelFinancial ReportingReconciliationAudit TrailsBudget AnalysisCost Centre MgmtVariance AnalysisCash Flow ModellingW-2 PreparationIFRS / GAAPGeneral LedgerMonth-End Close
Business Intelligence
BI & Visualization
Power BI92%
Tableau88%
Advanced Excel96%
Matplotlib / Seaborn85%
Database
SQL & Data Engineering
Advanced SQLWindow FunctionsQuery OptimisationETL PipelinesJoins & AggregationData CleaningBigQueryStar Schema
Operations & Strategy
Business Operations
Six roles across accounting, finance, operations, and analytics — backed by four qualifications. Strategy + domain + Python = a rare complete package.
Process MappingSOP DevelopmentWorkflow AutomationKPI MonitoringCross-Functional Coord.Stakeholder CommsData StorytellingFinancial Storytelling
04 — Work

Projects that
prove the point

Diabetes Prediction — Model Accuracy (100K patients · 2015–2018)
Neural Net
0.87
Grad. Boost
0.86
Rand. Forest
0.85
SVM
0.83
Log. Reg.
0.82
Dec. Tree
0.78
90,601 Non-Diabetic · 8,499 Diabetic · SMOTE · 5-Fold GridSearchCV
Machine Learning · Healthcare · Featured
Diabetes Dataset
Prediction Model
The Problem
Patient data scattered across silos — no model to flag high-risk individuals before costly hospitalisations.
The Outcome
Neural Net hit 0.87 AUC. HbA1c & blood glucose identified as top predictors, enabling early intervention at scale.
Python Scikit-Learn Pandas SMOTE GridSearchCV Neural Network Random Forest
↓ Download Report GitHub →
🩺
Project Preview
Case Study
Diabetes Prediction &
Healthcare ML Analysis
100K
Patients
0.87
Best AUC
87%
Neural Net
The Problem
Healthcare providers lacked a predictive tool to identify at-risk diabetic patients early. Data spanning Alabama & Wyoming (2015–2018) existed in silos with no consolidated model.
The Outcome
Neural Network topped at 0.87 AUC. HbA1c (0.49 importance) and blood glucose (0.40) emerged as dominant predictors — enabling preventative care decisions at scale.
My Approach
Built a six-model binary classification pipeline on 100,000 records. Applied SMOTE to address the severe class imbalance. Used 5-fold GridSearchCV to tune each model for optimal recall on the minority class.
Model Accuracy · SMOTE Balanced · 5-Fold GridSearchCV
Neural Net
0.87
Grad. Boost
0.86
Rand. Forest
0.85
SVM
0.83
Log. Reg.
0.82
Dec. Tree
0.78
PythonScikit-Learn PandasSMOTE GridSearchCVNeural Network Random ForestHealthcare Analytics Binary Classification
Resources & Downloads
↓ Download Report PDF
GitHub → LinkedIn →
Email About This Project
Payroll Anomaly Scanner — Audit Flags (724 records · 120 employees)
Excessive OT
12
Duplicates
8
Retro Adj.
8
Code Mismatch
6
Missing Ded.
2
Oracle Cloud HCM · 5 anomaly types · $130K+ irregular payroll exposure flagged
Python · Payroll Audit · Featured
Payroll Analytics
Pipeline
The Problem
Manual payroll review made it hard to catch duplicate pay, overtime abuse, and coding errors before payroll closed.
The Outcome
Python audit rules flagged 36 high-risk anomalies and surfaced $130K+ in irregular payroll exposure for review.
Python Pandas Oracle HCM Audit Rules Payroll QA Automation Anomaly Detection
GitHub →
💼
Project Preview
Case Study
Payroll Audit &
Analytics Pipeline
724
Records
120
Employees
$130K+
Exposure
The Problem
Payroll teams needed a faster way to identify duplicates, retro anomalies, coding mismatches, and deduction issues before payments were finalized.
The Outcome
A Python-led audit pipeline caught 5 anomaly categories and prioritized the highest-risk issues for human review, reducing manual checking time and missed exceptions.
My Approach
Built a rules-based payroll QA workflow over Oracle Cloud extracts, using Pandas transformations and exception logic to isolate excessive overtime, duplicate entries, retro adjustments, code mismatches, and missing deductions.
Audit Flag Counts · Payroll QA Pipeline
Excessive OT
12
Duplicates
8
Retro Adj.
8
Code Mismatch
6
Missing Ded.
2
PythonPandas Oracle HCMPayroll Controls AutomationAudit Pipeline Exception Detection
Resources & Links
GitHub →
Email About This Project
Revenue KPI Dashboard — Performance Metrics ($50M+ tracked)
QoQ Growth
14%
Collection
97%
ROI Mult.
4.2x
Forecast
94%
Leakage
$2M+
SQL + Tableau + Power BI · Weekly reporting automated to 24-hour turnaround
BI · Revenue Intelligence · Featured
Revenue Analytics
Dashboard
The Problem
Leadership needed faster visibility into revenue trends, leakage, and forecast confidence across high-value accounts.
The Outcome
Built live KPI dashboards tracking $50M+ in revenue, with 94% forecast accuracy and $2M+ leakage surfaced.
Tableau Power BI SQL Forecasting KPI Reporting Leakage Analysis
GitHub →
📈
Project Preview
Case Study
Revenue Intelligence &
Executive Dashboarding
$50M+
Revenue
94%
Forecast
$2M+
Leakage
The Problem
Revenue reporting was too slow and fragmented to support fast commercial decisions, especially around leakage and collection performance.
The Outcome
Interactive dashboards condensed operational and financial signals into one decision layer for executives and revenue teams.
My Approach
Modeled KPI pipelines in SQL, surfaced visual reporting in Tableau and Power BI, and tracked trend movement, collection efficiency, ROI, and forecast quality in near real time.
Revenue KPI Snapshot
QoQ Growth
14%
Collection
97%
ROI Mult.
4.2x
Forecast
94%
Leakage
$2M+
TableauPower BI SQLForecasting Executive ReportingRevenue Leakage
Enterprise Data Warehouse — Star Schema Coverage
Fact Table
100%
Dim: Time
72%
Dim: Cust.
56%
Dim: Prod.
46%
SQL · ETL · BigQuery · Self-service reporting foundation
SQL · Data Engineering · Featured
Enterprise Data
Warehouse
The Problem
Reporting depended on scattered source data and repeated manual joins, making consistent BI difficult for non-technical stakeholders.
The Outcome
Designed a full star-schema warehouse that simplified analytics and enabled reliable self-service reporting.
SQLStar SchemaETLBigQuery
GitHub →
🗄️
Project Preview
Case Study
Enterprise Warehouse &
BI Foundation
1
Fact Table
3
Dimensions
ETL
Pipeline
The Problem
Operational data was not structured for analytics consumption and required repetitive preparation before reporting.
The Outcome
A warehouse layer standardized grain, keys, and dimensions for scalable reporting and downstream dashboards.
My Approach
Modeled fact and dimension tables with surrogate keys, clear grain definitions, and ETL logic that supported clean BI outputs for business users.
Expense Fraud Flags — 1,240 Claims Screened
Duplicate
18
Round Amts
13
Split Inv.
10
Vendor Mis.
7
Holiday
4
Python anomaly detection · 5 fraud patterns · 62% fewer false approvals
Controls Analytics · Fraud Detection · Featured
Expense Fraud
Detection
The Problem
Expense review needed a faster way to highlight suspicious claims before approval and reimbursement.
The Outcome
Flagged high-risk claims across five fraud patterns and improved exception review quality with fewer false approvals.
PythonIsolation ForestPandasRules Engine
GitHub →
🧠
Project Preview
Case Study
Expense Controls &
Fraud Screening
1,240
Claims
5
Patterns
62%
Reduction
The Problem
Manual expense checks struggled to consistently identify suspicious duplicates, splits, and mismatches.
The Outcome
Automated anomaly checks elevated risky claims for review and strengthened fraud-prevention controls.
Cash Flow Forecast — 12 Month Horizon
Best Case
$4.2M
Base Case
$2.8M
Worst Case
$0.9M
Accuracy
91%
Burn Cover
7.4m
Scenario planning across AR, AP, and payroll cycles
Financial Modeling · Forecasting · Featured
Cash Flow
Forecasting
The Problem
Cash planning needed a forward-looking model that could support multiple scenarios instead of static spreadsheet snapshots.
The Outcome
Built best, base, and worst-case forecasts with 91% accuracy across an 18-month back-test.
PythonPandasExcelScenario Modeling
GitHub →
💵
Project Preview
Case Study
Cash Flow Planning &
Scenario Forecasting
12
Months
91%
Accuracy
3
Scenarios
The Problem
Leadership needed scenario-based cash visibility for near-term and medium-term planning.
The Outcome
The model linked operating cycles to liquidity projections and improved planning confidence.
Credit Risk Segments — 3,500 Accounts Scored
Low Risk
62%
Medium
24%
High Risk
11%
Default
3%
Logistic regression + gradient boosting · Default probability visualized in Power BI
Risk Modeling · Power BI · Featured
Credit Risk Scoring
Dashboard
The Problem
Credit teams lacked a clear ranking of customer default probability and needed a faster way to segment accounts by risk.
The Outcome
Scored 3,500 accounts across four tiers and exposed default likelihoods in a live dashboard for decision support.
Python Gradient Boosting Power BI Risk Segmentation SQL Default Prediction
GitHub →
🏦
Project Preview
Case Study
Credit Risk Modeling &
Portfolio Segmentation
3,500
Accounts
4
Tiers
3%
Default
The Problem
Account-level credit monitoring needed better prioritization than manual review and static rules could provide.
The Outcome
A modeling layer combined statistical scoring with executive-friendly risk visuals, improving visibility on low, medium, high, and default segments.
My Approach
Combined logistic regression and gradient boosting outputs, segmented the portfolio into action-ready bands, and surfaced account risk positions inside a Power BI dashboard for operational use.
Risk Segment Distribution
Low Risk
62%
Medium
24%
High Risk
11%
Default
3%
PythonGradient Boosting Power BIPortfolio Analytics SQLDefault Risk
Budget vs Actual — Q1 Variance Monitoring
Revenue
+3%
Payroll
-8%
OpEx
-3%
Marketing
+16%
Net Margin
-5%
Automated variance narratives · Python + Excel + Power BI · 70% less manual commentary
Finance Automation · Variance Analysis · Featured
Budget Variance
Intelligence
The Problem
Finance teams spent too much time writing repetitive commentary to explain monthly budget-to-actual movements.
The Outcome
Automated variance analysis generated plain-language explanations of movement drivers and cut manual commentary time by 70%.
Python Excel Power BI Variance Analysis Automation Narrative Reporting
GitHub →
🧾
Project Preview
Case Study
Budget-to-Actual &
Variance Commentary
70%
Time Saved
24h
Turnaround
5
Key KPIs
The Problem
Budget reviews were accurate but too slow because commentary and explanation of movements had to be prepared manually every cycle.
The Outcome
The workflow automated both detection and explanation of major variances, helping leadership understand impact faster.
My Approach
Combined Python and Excel-based variance logic with dashboard output in Power BI, then translated metric changes into concise natural-language commentary aligned to business reporting cycles.
Q1 Variance Snapshot
Revenue
+3%
Payroll
-8%
OpEx
-3%
Marketing
+16%
Net Margin
-5%
PythonExcel Power BIVariance Analysis NarrativesFinance Automation
Workforce Cost Analysis — 380 Employees
Overtime
22%
Absentee.
7.4%
Productiv.
82%
Cost/Output
12%
Savings
$85K
HR analytics + payroll + productivity monitoring
HR Analytics · Cost Optimization · Featured
Workforce Cost
Optimization
The Problem
Leaders needed a clearer view of where overtime, absenteeism, and staffing inefficiencies were driving unnecessary cost.
The Outcome
Analysis across 380 employees surfaced $85K in addressable savings and informed staffing recommendations.
PythonPower BIOracle HCMHR Analytics
GitHub →
👥
Project Preview
Case Study
Workforce Analytics &
Staffing Efficiency
380
Employees
$85K
Savings
82%
Output
The Problem
Payroll and workforce signals needed to be translated into practical cost-saving actions.
The Outcome
Analysis identified efficiency opportunities across overtime, absence, and productivity trends.
Revenue Leakage — Root Cause Drill-Down
Billing
620K
Discounts
480K
Write-offs
390K
Collections
310K
Recovered
1.8M
SQL + Tableau + Python · Four leakage categories isolated
Revenue Ops · Root Cause Analysis · Featured
Revenue Leakage
Analyzer
The Problem
Revenue leakage was occurring across several categories without a single view of root causes and recovery opportunities.
The Outcome
Identified and helped recover $1.8M by drilling down into billing, discounts, write-offs, and collection gaps.
SQLTableauPythonRevenue Ops
GitHub →
💸
Project Preview
Case Study
Revenue Recovery &
Leakage Analysis
$1.8M
Recovered
4
Drivers
SQL
Pipeline
The Problem
Leakage patterns were distributed across multiple operational processes and difficult to quantify together.
The Outcome
Centralized analysis exposed the highest-value recovery opportunities and supported action across four categories.
05 — Experience

Six roles.
One mission.

01
Payroll Specialist
Payroll & Compensation Audit
  • Owned multi-entity payroll execution for 500+ employees in Oracle Cloud HCM
  • Built anomaly detection workflows that surfaced $130K+ in irregular pay risk
  • Led bi-weekly controls across duplicates, overtime exceptions, and code mismatches
  • Automated payroll reporting packs, reducing manual review effort by 60%
  • Managed benefit deduction reconciliation and audit-ready year-end reporting
  • Maintained payroll tax compliance across multi-state requirements
02
Business Analyst
Business Analysis & Strategic Insights
  • Converted business requirements into measurable analytics and reporting solutions
  • Performed root-cause analysis to isolate process inefficiencies and cost leakage
  • Produced executive reporting that unified financial and operational performance
  • Defined strategy-aligned KPIs with cross-functional stakeholder ownership
  • Delivered gap analysis and feasibility recommendations for new initiatives
  • Served as translator between technical teams and business leadership
03
Finance Analyst
Financial Analysis & Reporting
  • Prepared monthly P&L, balance sheet, and cash flow packs for leadership review
  • Delivered budget-to-actual variance analysis with decision-ready commentary
  • Built Excel finance models using Power Query, pivots, and dashboard views
  • Supported audit readiness through disciplined reconciliations and documentation
  • Tracked cost-center over/under-spend and escalated actionable trends
  • Recommended margin improvement actions that increased operating margin by 8%
04
Revenue Analyst
Revenue Analytics & Intelligence
  • Built Tableau and Power BI revenue dashboards tracking $50M+ in performance
  • Automated financial reporting from weekly cycles to near real-time with SQL and Python
  • Deployed forecasting models achieving 94% accuracy across six-month horizons
  • Identified and quantified $2M+ in leakage through reconciliation analytics
  • Delivered CFO-level variance and trend insight for strategic planning
  • Reduced reporting cycle time from one week to 24 hours
05
Operations Specialist
Business Operations & Process Excellence
  • Redesigned workflows that reduced cross-department cycle times by 35%
  • Authored SOPs adopted across finance, HR, and operations teams
  • Implemented KPI monitoring dashboards for executive operations visibility
  • Coordinated Oracle Cloud migration activities across finance and payroll
  • Led process optimization initiatives with measurable cost savings impact
  • Managed simultaneous delivery across four or more functional departments
06
Accountant
Accounting & Financial Control
  • Managed full general ledger, journals, and month-end close operations
  • Performed accurate reconciliations across bank, AP, and AR ledgers
  • Ensured IFRS and GAAP compliance across financial reporting outputs
  • Maintained fixed asset registers and depreciation scheduling controls
  • Prepared trial balances, supporting schedules, and audit working papers
  • Shortened month-end close by three days through automation and control design
06 — Resume

A concise professional
dossier

Downloadable Profile
Lesley Shiri
Resume Snapshot

Built for recruiters and hiring managers who want a fast, credible read on fit: a concise summary of finance rigor, analytical judgment, reporting strength, and cross-functional execution across payroll, revenue, operations, and business analysis.

Download Resume PDF Discuss Opportunities LinkedIn Profile
What You Get
Finance + analytics: accounting rigor paired with dashboards, controls, modeling, and business-facing insight delivery.
Execution range: payroll controls, budgeting, revenue intelligence, forecasting, KPI reporting, and process improvement.
Decision support: work designed to improve clarity, reporting speed, operational control, and decision quality.
Best Fit Roles
Business Analyst for finance, operations, and reporting-heavy teams.
Finance / Revenue Analyst for forecasting, KPI reporting, variance review, and performance tracking.
Data Analyst focused on controls, dashboards, automation, and operational insight generation.
At A Glance
8+Years Experience
4Qualifications
6Professional Roles
3Core Domains

Direct contact: lesleylonelyshiri88@gmail.com. The resume PDF provides a cleaner, formal summary of the same experience presented throughout the portfolio.

Business Analytics Financial Reporting Python Power BI SQL Revenue Analysis Payroll Controls Forecasting
07 — Tech Stack

Tools of the trade

🐍
Python
📊
Power BI
📈
Tableau
🔢
Excel
🗄️
SQL
🤖
Scikit-Learn
🧮
Pandas
☁️
Oracle Cloud
🌲
Rand. Forest
⚡
Grad. Boost
🧠
Neural Nets
🎓
MBA · MCom
08 — Let's Build
READY TO
WORK TOGETHER?

Open to business analyst, finance analyst, revenue analyst, and data analyst opportunities where strong reporting, control, and analytics execution are valued.

Directlesleylonelyshiri88@gmail.com
Email Me Download Resume LinkedIn
GitHub Project Report
Download Resume