Home › Projects
Portfolio

Example Projects & Case Studies

Real analyses on real datasets — showing exactly the kind of work I deliver to clients.

1
In-depth analysis
A rigorous statistical coding analysis covering data cleaning, modeling, and visualization — every finding backed by the numbers.
2
Full written report
A professional PDF report translating the findings into plain English — with specific, actionable recommendations your team can act on immediately.
3
Interactive dashboard
A Power BI-style dashboard that brings the key insights to life — no software license required, built to be shared with any stakeholder.
Example Project 01
Sales & Customer Performance — UK Online Retailer

Full end-to-end analysis of 541,910 transactions from a UK-based online retailer covering December 2009 through December 2011 — product performance, seasonal demand, geographic opportunity, pricing efficiency, and customer value.

Statistical Coding Regression Modeling Retail Analytics Data Cleaning Dashboard
Key findings
80/20
Top 20% of products drive 80% of total revenue
Z=2.72
November spike significant at 99% confidence level
2×
International customers spend twice as much per order
£1.5M+
Combined lifetime spend across top 10 customers
Resources
Dataset
UCI Online Retail II — Kaggle
Source code
View full analysis notebook
Full report
Download PDF report
Live interactive dashboard
krenzanalytics.com — Retail Sales Dashboard
Total revenue
£10.44M
Dec 2009 – Dec 2011
Total orders
22,190
Unique invoices
Unique products
3,919
Top 20% → 80% revenue
Unique customers
4,338
135,080 txns missing ID
Monthly revenue trend
November 2011 peak (Z=2.72) — plan Q4 inventory early
Revenue by country
UK leads; intl orders 2× avg value
Top products
Protect these SKUs from stockouts
Top 10 customers by revenue
High-value accounts — prioritize for retention
RankCustomer IDRevenueOrdersAvg order
114646£280,20677£3,639
218102£256,438145£1,768
317450£194,55155£3,537
416446£168,47228£6,017
514911£143,825201£715
612415£124,91487£1,436
714156£117,37972£1,630
817511£91,062115£792
916029£81,02463£1,286
1012971£77,18395£813
Pricing efficiency — over & under performers
Products >2 SD from predicted sales volume
Over-performers
ProductPrice
Ceramic storage jar£1.25+over
WW2 gliders£0.21+over
Jumbo bag retrospot£1.95+over
White heart t-light£2.95+over
72pk cake cases£0.55+over
Under-performers
ProductPrice
Cakestand 3 tier£12.75-under
Rabbit night light£34.95-under
Umbrella stand£22.50-under
Xmas tree large£19.95-under
Paper craft birdie£2.95-under
Krenz Analytics  |  Source: UCI Online Retail II  |  Diego Krenz, May 2026
Example Project 02
Customer Churn Analysis — Telecom Provider

Full end-to-end churn prediction analysis on 7,043 customer records. Logistic regression model with backward elimination, risk tiering, and $236K annual revenue impact quantification.

Python Logistic Regression Customer Analytics Scikit-learn Dashboard
Key findings
42.7%
Month-to-month churn rate vs 2.8% on two-year contracts
0.838
AUC-ROC — model ranks churner above non-churner 84% of the time
53%
New customers churn in first 6 months — critical retention window
$236K
Estimated annual revenue protected at 30% retention rate
Resources
Dataset
IBM Telco Customer Churn — Kaggle
Source code
View full analysis notebook
Full report
Download PDF report
Live interactive dashboard
krenzanalytics.com — Telco Churn Dashboard
Customers
7,043
21 variables
Churn Rate
26.5%
1,869 churners
AUC-ROC
0.838
Logistic regression
Annual Savings
$236K
30% retention
Churn by Contract Type
Month-to-month churns 15x two-year rate
Churn by Tenure Group
First 6 months: 53% churn rate
Churn by Payment Method
Electronic check 3x auto-pay churn rate
Risk Tiers
Based on predicted churn probability
TierProbCountAction
High>70%496Immediate outreach
Med40-70%1,521Proactive check-in
Low<40%392Standard service
Annual savings~$236K
Krenz Analytics  |  IBM Telco Customer Churn  |  Diego Krenz, May 2026
View full project
Ready to understand your data?
Tell me what you're working with and I'll come back with what's possible.
Get in touch