GUIDED PROJECT BEGINNER DIAGNOSTIC

EDA & Storytelling what drove performance last year?

The CMO wants to understand what drove last year's best and worst months. You'll run an end-to-end EDA on a real UK retail dataset, find the drivers, and turn the analysis into a 6-slide story for the board.

TOOLS & TECHNIQUES

Python pandas EDA Matplotlib Storytelling Stakeholder writing

▶ Start chapter 1

Google Meta

I've personally interviewed hundreds of data candidates.

Pasquale Bruno Founder · datascienceportfol.io Read bio

⏱

5–7 hours

self-paced

◉

7 chapters

video + written

◱

Real dataset

UCI Online Retail · 541k rows

✦

Built with Claude

Ch 5: narrative copy

→

1-click publish

to your portfolio

WHAT YOU'LL WALK AWAY WITH

Real, portfolio-ready artifacts, not a certificate.

✓

Full EDA notebook

✓

6-slide stakeholder deck

✓

Written findings summary

✓

Portfolio case study page

THE SCENARIO

Real CMO question. Real retail data. Real board deck.

You're a data analyst at a UK online retailer. The CMO wants to know **what drove our best and worst performing months last year**, something they can take into the board meeting.

You have **541k transactions** from Dec 2010 to Dec 2011 (UCI Online Retail dataset). Your job: run the EDA, isolate the drivers, and turn the analysis into a **6-slide stakeholder deck** the CMO can actually present.

BY THE END, YOU'LL BE ABLE TO

✓

Frame an open-ended business question as a structured EDA

✓

Use pandas + Matplotlib to find revenue drivers across products, geographies and time

✓

Layer Claude API to generate written narrative copy from chart data

✓

Build a stakeholder deck that survives 4 seconds of exec attention

✓

Ship a polished GitHub repo + LinkedIn post that reads like real analyst work

PROJECT CURRICULUM

7 chapters · ~5–7 hours

CLAUDE API NEW 14 videos 7 written lessons

Project brief & setup

video + setup

35 min +

● Walk through the brief: a Monday-morning Slack message from your manager, board meeting Thursday

● Meet the dataset: UCI Online Retail · 541k transactions · Dec 2010–Dec 2011

● Set up Google Drive, GitHub repo, and Colab; verify the CSV loads with the right encoding

● Write a pre-analysis plan: business question, hypotheses, and what would surprise you, before touching any code

Meet the data

video + feature engineering

50 min +

● First look: shape, nulls, and the 25% CustomerID gap (wholesale orders with no account)

● Engineer `revenue` (Quantity × UnitPrice) and a 6-category product taxonomy from 4,000+ descriptions

● Understand cancellations (1.7% of rows) without dropping them yet

● Chart 1: transaction volume by month, and why December looks artificially low

Revenue trends

video + EDA

55 min +

● Gross vs net revenue, and why the ~£900k gap matters for an honest number

● Find the best month (Nov 2011) and worst month (Apr 2011): a 2.4x swing

● Chart 2 (the main chart of the project) plus weekly-pattern and rolling-average charts

● Investigate the top 3 revenue days and catch a single £50k wholesale invoice masquerading as a demand spike

What drove performance

video + EDA

1h 5m +

● Break performance down 3 ways: category, country, and order size

● Chart 4b: 100% stacked category mix proves Seasonal & Festive drove the Q4 swing, not general uplift

● UK accounts for ~83% of revenue: a concentration risk and a growth opportunity

● Order value distribution on a log scale reveals the wholesale customer tail

The returns story

video + data quality

55 min +

● Data quality catch: AMAZON FEE, Manual and POSTAGE were sitting in cancellations, inflating one category's return rate

● Go back and fix Ch 3. This is what real analysis looks like

● Chart 8 & 9: monthly cancellation rate and return rate by category (Home & Decorative highest at 4.6%)

● Reframe the best month: November's 3.2% cancellation rate is well below the 9.3% annual average, so the revenue was clean

Build the stakeholder deck 🧠 CLAUDE API

video + Claude API

1h 20m +

● Select 4 charts from 12. The analytical-vs-communication judgment call is yours, not Claude's

● One Claude call per slide: turn pre-computed findings into CMO-ready prose

● Build a 6-slide deck end-to-end with python-pptx

● Review and edit every Claude-drafted sentence before it ships

Ship it + frame for employers

video + ship

50 min +

● Draft the README with Claude, Limitations and Next Steps sections included

● Push the complete repo to GitHub: notebook, clean CSV, 11 charts, deck, README

● Draft the LinkedIn post: lead with the finding, link in the first comment

● The STARL interview framework + the 3 questions you'll actually be asked

▶ Start chapter 1

PREVIOUSLY

Google Meta

YOUR INSTRUCTOR

Pasquale Bruno Founder · datascienceportfol.io

I spent 10+ years in data and product roles at Google and Meta, including hundreds of hiring-panel interviews, so I know exactly what makes a hiring manager keep reading. I design and review every guided project myself, using briefs and datasets modelled on real analyst work, not toy CSVs. If you get stuck, you're emailing me directly, not a support queue.

10+ yrs

IN INDUSTRY

500+

CANDIDATES INTERVIEWED

300+

PORTFOLIOS REVIEWED

See all projects Connect on LinkedIn

SKILLS YOU'LL PRACTICE

Tools & techniques

Python pandas EDA Matplotlib Storytelling Stakeholder writing Feature engineering Log-scale distributions Returns analysis Slide design

PREREQUISITES

Before you start

Comfortable with Python basics (variables, functions, loops) REQUIRED

Familiar with pandas DataFrames (groupby, merge) REQUIRED

Some Matplotlib or similar for charts HELPFUL

Familiarity with retail / e-commerce business model HELPFUL

🧠 NEW · USE AI LIKE A REAL JOB

Ship analysis with the Claude API in the loop.

Real data teams aren't writing SQL one query at a time anymore. A whole chapter teaches you to wire Claude into a real analysis, the way analysts at Stripe, Anthropic and Notion actually work.

Generate slide narrative copy

One Claude call per slide, written from your actual chart data

Stakeholder-ready prose

No more 'data → bullet points'. Claude writes the why

Reusable pattern

The same one-call-per-chart approach works on every EDA you'll do

Pricing

One plan gets you every Guided Project.

Free builds your portfolio. Job Ready adds every Guided Project on top, live now and every one we ship next.

Free

$0/forever

For anyone with projects ready to share.

✓ Portfolio website with unlimited projects
✓ Personalised URL
✓ No design nor technical skills needed
✓ All your education and experience info in one place

Start free

Job Ready

$99/ month

For people investing in a data career.

✓ Portfolio website builder PRO
✓ All current + future Guided Projects
✓ Discord support channel for any questions, answered by me
✓ Weekly live office hours with me
✓ 1:1 monthly call with me (first 20 members)
✓ Async portfolio & LinkedIn reviews (first 20 members)
✓ Cancel anytime

Get Job Ready

Not sure if it's right for you?

Book a free 15-minute call with me. We'll talk through where you are and whether a Guided Project makes sense.

Book a call →

HOW WE'RE DIFFERENT

Why this isn't another course.

FEATURE THIS PROJECT DATACAMP COURSERA

Real business scenario, not toy data ✓ ✕ ✕

Portfolio-ready deliverable ✓ ✕ –

1-1 feedback on your writeup ✓ ✕ ✕

Published to your own site ✓ ✕ ✕

Certificate of completion ✓ ✓ ✓

Thousands of micro-lessons ✕ ✓ ✓

FREQUENTLY ASKED

Questions, answered.

Do I need advanced Python? +

No, basic pandas (DataFrames, groupby) is enough.

Can I cancel anytime? What about refunds? +

Yes, cancel anytime, no lock-in, right from your account. We don't offer refunds on subscription payments already made, but you'll keep full access through the end of the billing period you've already paid for.

Is the certificate worth anything? +

It's a signal, not a credential. The writeup is what gets you hired.

Can I use this in interviews? +

Yes, and we encourage it. The framework is reusable across case-study rounds.

CONTINUE YOUR PATH