Skip to content

Université Catholique de Lille · M2 International Management

Data Analytics for Business

Welcome. This is the companion page for your 2026-2027 course at Université Catholique de Lille (la Catho). It mirrors the syllabus and points you to free, browser-based practice you can do before and after each session, no installs and no account needed.

University students collaborating on a laptop outside a campus building
Made for the M2 International Management cohort.
Photo : George Pak · Pexels

By Sitraka Forler · Lecturer, Durham Business SchoolUpdated 13 September 2026 About this site

18 hours

6 sessions · Dec 2026 – Jan 2027

English

no coding background assumed

Campus Vauban

5 on campus · 1 remote

One case

Olist e-commerce, ~100k orders

What you will be able to do

  1. 1Frame a business problem as an analytical question and pick the right type of analytics (descriptive, diagnostic, predictive, prescriptive) and KPIs.
  2. 2Clean real tabular data and retrieve it yourself with core SQL (SELECT, WHERE, GROUP BY, JOIN).
  3. 3Design honest, decision-oriented visualisations and dashboards, and tell the story from "what" to "so what" to "now what".
  4. 4Interpret regression, classification and clustering outputs, evaluate them with business-relevant metrics, and spot overfitting, leakage and biased samples.
  5. 5Use generative AI as an assistant: prompt well, verify outputs, and disclose how you used it.
  6. 6Judge the governance of data and AI in Europe: GDPR essentials, the EU AI Act, and the manager’s accountability for AI-assisted decisions.
A team analysing business data on printed charts and a laptop
The whole point: turning data into a decision.
Photo : Yan Krukau · Pexels

The six sessions, with practice

Every practice link below is free and runs in your browser. The lessons finance-flavour the examples, but the skills are the same ones the course teaches on the e-commerce case.

1

From Data to Decisions

Wed 9 Dec 2026 · 4h · On campus

The analytics maturity ladder (descriptive → diagnostic → predictive → prescriptive), data literacy, KPI trees, and the cost of dirty data. Hands-on cleaning and pivots on the Olist case; groups get their management brief.

Go deeper on EcoFinLearn

What machine learning really is (the four analytics types, in plain terms)Cleaning messy data: duplicates, missing values, outliersCorrelation, the first diagnostic question
2

Data Visualisation & Storytelling

Thu 17 Dec 2026 · 3h · Remote

Chart choice and perception, decluttering and honest axes, dashboards for executives, and narrative structure. Each group builds a live Looker Studio dashboard and critiques another’s.

Go deeper on EcoFinLearn

The interactive explainers: worked examples of honest, decision-first chartsReading a correlation matrix visuallyLooker Studio (build the dashboard)
3

SQL Essentials for Managers

Thu 7 Jan 2027 · 2h · On campus

The relational model in twenty minutes, then SELECT, WHERE, ORDER BY, GROUP BY/HAVING and JOINs across the case tables, plus reading and challenging queries written by others (including by AI).

Practise before class (free, in your browser)

SQL Foundations, the whole trackSELECT, WHERE and ORDER BYGROUP BY and aggregatesJOINs across two tablesDuckDB in-browser shell (run SQL on the case)
4

Predictive Analytics for Decision-Makers

Thu 7 Jan 2027 · 3h · On campus

Regression, classification and clustering as business tools; train vs test, overfitting and leakage; and the metrics that matter to a manager, the cost of a false positive vs a false negative, not just accuracy.

Go deeper on EcoFinLearn

Overfitting and cross-validationLinear regression, from the coefficient upClassification with logistic regressionThe confusion matrix and the cost of being wrongOrange Data Mining (no-code ML)
5

Generative AI, Governance & the EU AI Act

Fri 8 Jan 2027 · 3h · On campus

LLMs as analytical assistants, what they do well and where they fail, and verification workflows. GDPR essentials, the EU AI Act’s risk-based architecture and live timeline, and what "deployer" obligations mean for a manager.

Go deeper on EcoFinLearn

EU AI Act, the European Commission’s official portalThe AI Assessment Scale (AIAS), the framework grading AI use in this courseWhen deployed models rot: data drift as a governance concern
6

Capstone - Pitch Your Insights

Fri 8 Jan 2027 · 3h · On campus

Each group delivers a 10-minute executive pitch: one clear recommendation for its brief, backed by the dashboard and analysis built across the course, then 5 minutes of Q&A. Plus a short individual quiz.

Go deeper on EcoFinLearn

A worked end-to-end capstone: a data-backed research reportWhere analytics roles go next, by function and region

Assessment and the AI rules

50%

Group capstone

dashboard, a 4-page brief, and the live pitch with Q&A

30%

Individual quiz

30-min closed-book, in Session 6

20%

Milestones & participation

cleaned data, draft dashboard, model memo

AI fluency is a learning outcome here, not a threat, but every piece of work is labelled with a level of the AI Assessment Scale. Honest disclosure is never penalised; undisclosed AI where it is not allowed, or fabricated data or sources, is misconduct. You are accountable for everything you submit: "the AI said so" is not a defence here, or later under the EU AI Act.

1 · No AISupervised, AI-free: the individual quiz and the live pitch and Q&A.
2 · AI for planningAI may help you brainstorm or structure ideas; the work is your own.
3 · AI for editingAI may polish language or layout of work you produced; substance stays yours.
4 · AI + human evaluationAI may help produce analysis if you check, correct, own and disclose every output.
5 · AI-integratedAI use is the exercise (Session 5): you are graded on how you direct and verify it.

The toolbox

All free, all in the browser. This is the stack you can reuse on day one of a job.

Google Sheets / Excel - cleaning & pivotsLooker Studio - dashboardsDuckDB shell - in-browser SQLOrange Data Mining - no-code MLOlist dataset (Kaggle) - the running case

Further reading

Where to go deeper. Free means a full, legal copy is online.

Frequently asked questions

Do I need any coding or maths background?

No. The course assumes none, and the practice linked here starts from zero: spreadsheets first, then SQL and no-code machine learning.

Is the practice on this page free?

Yes. Every lesson and tool linked here is free, runs in your browser, and needs no account or install.

Which tools does the course use?

All free and browser-based: Google Sheets or Excel, Looker Studio, DuckDB for SQL, Orange for no-code machine learning, and a general AI assistant, all on the Olist e-commerce dataset.

Is the course taught in English?

Yes, the sessions and assessments are in English. This page is also available in French to help you prepare.

How is the course graded?

50% group capstone (dashboard, brief and live pitch), 30% individual quiz, and 20% milestones and participation. Every piece of work is labelled with an allowed level of AI use.

Can I use ChatGPT or Claude for the work?

Yes, transparently. Each task specifies an allowed level on the AI Assessment Scale, and honestly disclosing how you used AI is never penalised.

New here? Start with SQL

It is the highest-leverage skill in the course and the fastest win. Ten minutes now makes Session 3 easy.

Run your first query, free

This companion is maintained by your instructor, Sitraka Forler, and is not an official publication of Université Catholique de Lille. The authoritative syllabus, dates and rooms are the ones on the university's learning platform. Spotted an error? tell me and I will fix it.