Skip to content

Université Catholique de Lille · M2 International Management · 2026–2027

Data Analytics for Business

Every function an international manager will lead now runs on data. This course does not turn students into data scientists: it prepares them to commission analytical work, question it and act on it.

  • 3 ECTS
  • 6 sessions, 18 hours, December 2026 to January 2027
  • Taught in English
  • Campus Vauban, Lille, and one session online

Where we are

The first session is on Wednesday 9 December 2026, 13:00 to 17:00, on campus. Dates and deadlines are listed below.

Times are Paris time. All dates

Sessions

  1. Session 1: From data to decisions

    Wednesday 9 December 2026 · 13:00–17:00 · on campus

    Before, during and after (4 hours)
    Before
    About 1.5 hours. Read Provost and Fawcett, chapters 1 and 2. Bring a laptop; the case files are handed out in class.
    In class
    The course contract (assessment, AI policy, groups). The analytics ladder, and why so many data initiatives fail. KPI trees: a worked example, then your turn. Cleaning the case data and first pivot tables. Groups are formed and receive their management briefs. A one-minute exit ticket, then the first quiz.
    After
    Milestone 1, due Tuesday 15 December at 23:59: your cleaned working dataset and a one-page KPI tree for your brief.
    Key concepts
    Analytics maturity, KPI, leading and lagging indicators, data quality, granularity, pivot table.
  2. Session 2: Data visualisation and storytelling

    Thursday 17 December 2026 · 13:00–16:00 · online · How to join

    Before, during and after (3 hours)
    Before
    About 2 hours. Read Knaflic, chapters 1 to 3. Follow the setup guide on this page to connect the case data to Data Studio, so that class time goes to design rather than to logins.
    In class
    Collective feedback on Milestone 1. How the eye reads a chart. A chart clinic on published charts, some of them misleading. In breakout rooms, each group builds a first dashboard for its brief. Peer critique and the five-second test. Quiz in the last ten minutes.
    After
    Milestone 2, due Monday 4 January at 23:59: your draft dashboard and a three-sentence narrative of your main finding. Written feedback by Wednesday 6 January.
    Key concepts
    Chart choice, pre-attentive attributes, clutter, honest axes, dashboard hierarchy, narrative arc.
  3. Winter break

    18 December 2026 to 6 January 2027 · about 12 hours per student

    Your group carries out the core of its analysis and drafts its dashboard and brief. A sensible plan: three blocks of about four hours, two before 24 December and one after 1 January. I answer emails until 22 December and again from 4 January, so send any questions about Milestone 2 before the 22nd.

  4. Session 3: SQL essentials for managers

    Thursday 7 January 2027 · 10:00–12:00 · on campus

    Before, during and after (2 hours)
    Before
    About 1 hour. Complete the 30-minute SQL warm-up sheet, and read the feedback on Milestone 2.
    In class
    How the case tables fit together. A live demonstration in DuckDB, in the browser. Exercises of increasing difficulty: complete a half-written query, then write one from a business question. Spot the bug in AI-generated queries that look right. Quiz.
    After
    Add one SQL-based finding to your brief.
    Key concepts
    Table, primary and foreign key, join, aggregation, row duplication.
  5. Session 4: Predictive analytics for decision-makers

    Thursday 7 January 2027 · 13:00–16:00 · on campus

    Before, during and after (3 hours)
    Before
    About 1.5 hours. Read Provost and Fawcett, chapters 5 and 7 (overfitting; what makes a model good).
    In class
    Intuition first, without formulas. A prepared Google Colab notebook: you run it, change a few settings and read the results; you do not write code. Groups compare two models with a simple cost table. You write Milestone 3 in class, then take the quiz.
    After
    Milestone 3 is handed in at the end of the session: a one-page memo per group on what your model predicts, how well, its limits and the decision you would or would not base on it.
    Key concepts
    Training and test data, overfitting, leakage, precision and recall, expected value, segmentation.
  6. Session 5: Generative AI, governance and the EU AI Act

    Friday 8 January 2027 · 09:00–12:00 · on campus

    Before, during and after (3 hours)
    Before
    About 1 hour, best done during the winter break. Read the European Commission's summary page on the AI Act, and bring your capstone draft.
    In class
    Part 1, about 70 minutes: an AI workshop on your own project. Part 2, about 70 minutes: GDPR essentials for analytics, the AI Act's risk tiers and its 2026 amendment, and a case exercise: is it high-risk? Part 3, about 30 minutes: capstone studio, then the last quiz.
    After
    Pitch slides and dashboard link are due on Friday 8 January at 12:45.
    Key concepts
    Verification, prompt patterns, hallucination, lawful basis, data minimisation, pseudonymisation, risk tiers, deployer, transparency obligation.
  7. Session 6: Capstone: pitch your insights

    Friday 8 January 2027 · 13:00–16:00 · on campus · Written exam and group pitches

    Before, during and after (3 hours)
    Before
    Be seated by 12:55 with a pen and your student ID. The corrections of the quizzes for Sessions 3 and 4 open on the morning of the exam, for revision.
    In class
    13:00 to 13:30: individual written exam, closed book, on paper. 13:35 to 15:35: group pitches in an order drawn at random, 15 minutes each (10 of pitch, 4 of questions, 1 to change over); every member answers at least one question. 15:35 to 16:00: debrief and next steps.
    After
    Final brief, AI-use appendix and peer-contribution form are due on Monday 11 January at 23:59.
    Note
    If there are more than eight groups, pitch time is shortened and the final timings are announced in Session 5.

For enrolled students: each session's slides open with the password given in class. Questions: [email protected].

Session 2 online: how to join

Thursday 17 December 2026, 13:00 to 16:00, Paris time. The session is live and interactive, and it is not recorded.

  1. The link to the video call is given in class in Session 1 and sent to you by email. It is not published on this page.
  2. Use a laptop, not a phone: you will build a dashboard in Data Studio during the session.
  3. Before the day, follow the setup guide and connect the case data to Data Studio with your Google account.
  4. Test your camera, microphone and connection, and join 10 minutes early, at 12:50.
  5. Keep your camera on during group work in the breakout rooms.
  6. The end-of-session quiz opens at 15:50, from this page.

Setup guide: Data Studio, PDF (3 pages)

A problem on the day: write to [email protected].

Deadlines

The January block is intensive: four sessions in two days. Plan your exam revision and the Session 5 reading before 7 January, so that the evening of 7 January is free to rehearse your pitch. The brief is due three days after the pitch, so that you can use the questions you received.

Table 1: Deadlines, Paris time.
WhatDue
Milestone 1: cleaned dataset and KPI treeLearning platformTue 15 Dec 2026, 23:59
Milestone 2: draft dashboard and three-sentence narrativeLearning platformMon 4 Jan 2027, 23:59
Milestone 3: model evaluation memo, one per groupWritten in classThu 7 Jan 2027, end of Session 4
Pitch slides and dashboard linkLearning platformFri 8 Jan 2027, 12:45
Written exam: be seated, with a pen and your student IDExam room, on campusFri 8 Jan 2027, 12:55
Final brief, AI-use appendix and peer-contribution formLearning platformMon 11 Jan 2027, 23:59
End-of-session quizzes (Table 3)This pageIn class; late windows

A late deliverable loses 1 point out of 20 on its own mark per started day, unless there is a documented reason. Talk to me before a deadline rather than after it.

Assessment

Every learning outcome is assessed. Grades are given on the French scale from 0 to 20.

Table 2: Assessment.
Assessment typeWeight
Continuous assessment
Milestones (group)15 Dec; 4 Jan; 7 JanAI use: Milestone 1: Level 2. Milestones 2 and 3: Level 315%
Participation, including the end-of-session quizzesSessions 1 to 6AI use: Quizzes: Level 1, open notes5%
Group project: dashboard and analytical briefDashboard 8 Jan, 12:45; brief of at most 4 pages 11 Jan, 23:59AI use: Level 3, with disclosure30%
Group pitch15 minutes, Session 6AI use: Preparation: Level 3. Delivery and questions: Level 120%
Final exam
Individual written exam, closed book30 minutes, Session 6AI use: Level 130%
Total100%

Open assessments, such as the project and the milestones, let you work the way you will at work, including with AI. Secure assessments, the written exam and the live questions after your pitch, check what you can do on your own. Individual marks for group work may be adjusted where the confidential peer-contribution forms show a clear imbalance, always after a conversation with the group.

End-of-session quizzes

After Sessions 1 to 5, a quiz of 10 questions in 10 minutes checks the key ideas; from Session 2 on, two questions come back from earlier sessions. Students take it on their own and without AI, with open notes, once, signed in with a Google account (or on paper, with the same weight). Participation credits taking the quiz in its window, not the score; the best four of five count. The full corrections open when the quiz closes.

Table 3: Quiz windows, Paris time.
QuizOpensCloses
Session 1Wed 9 Dec, 16:40Wed 16 Dec, 23:59
Session 2Thu 17 Dec, 15:50Wed 6 Jan, 23:59
Session 3Thu 7 Jan, 11:50Fri 8 Jan, 08:00
Session 4Thu 7 Jan, 15:55Fri 8 Jan, 08:00
Session 5Fri 8 Jan, 11:50Mon 11 Jan, 23:59

The quizzes for Sessions 3 and 4 close on the morning of the exam, so that their corrections can be used for revision.

How quiz data are handled

Generative AI policy

Using AI well is one of the outcomes of this course, so AI is allowed, within limits that vary from task to task. Every task carries a level from the AI Assessment Scale (Perkins, Roe and Furze, 2025). The levels describe what each task is designed for; they are not a detection tool. The use allowed widens as the course progresses, from Level 2 in December to Level 4 in the Session 5 workshop.

Table 4: AI Assessment Scale levels in this course.
LevelWhat it means
1. No AIKnowledge and skills are shown without AI, here in a supervised setting.Written exam; end-of-session quizzes; pitch delivery and questions
2. AI PlanningAI may support planning and the development of ideas; what you submit is your own work.Milestone 1
3. AI CollaborationAI may help complete the task, including drafting, feedback and refinement. You check every output and take responsibility for it.Milestones 2 and 3; group project; pitch preparation
4. Full AIAI involvement is expected: the goal cannot be reached by AI, or by a person alone, in the time available.Session 5 workshop
5. AI ExplorationAI is used creatively to solve problems or develop new solutions.Not used in this course
  • Disclosure. The group project includes a half-page appendix stating which AI tools you used, for which tasks, and how you checked their outputs. Honest disclosure is never penalised. Undisclosed AI use, AI use in a supervised assessment, and invented data, sources or results are breaches of academic integrity.
  • Responsibility. You are accountable for everything you submit. “The AI said so” is not an acceptable justification, in this course or at work.
  • Data protection. Never paste personal, confidential or proprietary data into a public AI tool. The course dataset is anonymised open data precisely so that you can practise without that risk.

Course notes, open to everyone

The slides are encrypted for the students. Each deck also has a public summary: the key ideas, the formulas and a worked example, in English, French and German.

All the notes, as a reading list

All 6 notes
  1. Session 1

    From data to decisions: the analytics ladder, KPI trees and honest data cleaning

    Frame a business question, break a goal into a KPI tree, and clean a real e-commerce dataset before you count anything: a manager's method.

  2. Session 2

    Data visualisation and storytelling: choosing the chart, fixing misleading ones, writing the finding

    How the eye reads a chart, how to pick the right one, four ways charts mislead, and how to turn a dashboard into a finding a manager can act on.

  3. Session 3

    SQL essentials for managers: reading queries, checking joins and auditing AI-written SQL

    Read and check SQL as a manager: keys and grain, SELECT, WHERE, GROUP BY, joins that duplicate or drop rows, and how to audit AI-written queries.

  4. Session 4

    Predictive analytics for decision-makers: what models learn, four traps, and the price of each error

    What a model learns, four traps (overfitting, leakage, biased samples, false causes), and how to price a confusion matrix to choose between two models.

  5. Session 5

    Generative AI, GDPR and the EU AI Act in 2026: what an analyst and a manager need to know

    Use an AI assistant on a data project and check its output, then the GDPR questions and the EU AI Act risk tiers and dates as amended in 2026.

  6. Session 6

    Pitching data insights to decision-makers: recommendation first, three pieces of evidence, honest answers

    A ten-minute structure for presenting a data-backed recommendation, what convinces an executive audience, how to handle questions and the usual mistakes.

Syllabus

The detailed course syllabus, version 1.2 of 1 November 2026, follows the structure of the faculty course sheet. Content may be adjusted before the course starts; the assessment structure will not change after Session 1.

Syllabus, PDF (14 pages)Setup guide: Data Studio, PDF (3 pages)

1. Pre-requisites

Good comprehension of English. Good knowledge of general business and management concepts. No prior background in statistics, programming or databases is required: all technical content is built up from the beginning, and every tool used in class runs in a web browser.

2. Course objectives

  • Develop the data literacy an international manager needs to commission, challenge and act upon analytical work, rather than to produce it as a specialist.
  • Go through the full analytics value chain, from framing a business question to defending a recommendation, on a single real international dataset.
  • Gain practical command of the core tools of business analytics: spreadsheets, dashboards, SQL queries and no-code predictive models.
  • Communicate quantitative findings persuasively to non-technical decision-makers.
  • Examine the governance, ethical and regulatory issues raised by data and AI in Europe, in particular the GDPR and the EU AI Act.
  • Take a critical and reflexive approach to data and to the outputs of generative AI tools, including their limitations and failure modes.

Why this course, now

  • Data literacy has become a baseline expectation. In DataCamp's 2026 survey of more than 500 enterprise leaders in the United States and the United Kingdom, 88% said basic data literacy is important for day-to-day work, yet 60% said they face a data skills gap.
  • Business schools increasingly treat AI literacy as a core skill. A framework on AI in business education published in January 2026 by Inspire Higher Ed with AACSB, GMAC and the Graduate Business Curriculum Roundtable observes that AI literacy is increasingly seen as a core competency rather than a specialisation.
  • In Europe it is also a matter of law. Since February 2025 the EU AI Act has required organisations that provide or use AI systems to take measures on their staff's AI literacy; an amendment in force since July 2026 frames this as supporting AI literacy rather than guaranteeing a given level. Most of the Act has applied since August 2026, and the obligations for high-risk uses such as AI-assisted recruitment will start to apply on 2 December 2027, when many of you will be in your first management roles.

3. What students learn to do

LO1
Frame. Translate a business problem into an analytical question, choose the type of analytics that fits it (descriptive, diagnostic, predictive or prescriptive) and define the indicators that will show progress.
LO2
Prepare and retrieve. Clean real-world tabular data in a spreadsheet, and retrieve what you need from a multi-table dataset with core SQL (SELECT, WHERE, GROUP BY, JOIN).
LO3
Design and communicate. Build honest, decision-oriented charts and dashboards, and structure a finding into a short, persuasive narrative.
LO4
Interpret models. Read the output of common predictive techniques (regression, classification, clustering), evaluate it with metrics that reflect business costs, and spot overfitting, data leakage, biased samples and the confusion of correlation with causation.
LO5
Use AI responsibly. Use a general-purpose AI assistant to speed up analytical work, prompting it effectively, checking its outputs and documenting how it was used.
LO6
Assess governance. Identify the GDPR and EU AI Act questions raised by an analytics or AI project, and explain what a manager remains accountable for when a decision is supported by AI.

4. Course content

  • The analytics maturity ladder: descriptive, diagnostic, predictive and prescriptive analytics.
  • Data literacy and data-driven decision-making: from strategic goals to KPI trees.
  • Data types, data quality and the business cost of poor data.
  • Data visualisation: perceptual principles, chart selection, dashboard design for executives.
  • Data storytelling: structuring an analysis into a recommendation for a non-technical audience.
  • The relational model and SQL essentials for managers.
  • Predictive analytics: regression, classification and clustering, training and test data, evaluation with business-relevant metrics, correlation and causation.
  • Generative AI as an analytical assistant: prompting patterns, verification workflows, hallucination and other limitations.
  • Data and AI governance in Europe: GDPR essentials for analytics, the risk-based architecture of the EU AI Act and its timeline, obligations of deployers of AI systems.

5. Learning and teaching methods

Lectures, guided demonstrations, hands-on workshops and a group project. Students are expected to take part actively in class and in group work. Groups of four or five are formed in Session 1, mixing backgrounds and nationalities on purpose. Students bring their own laptop to every session, including the online one.

One real case, from the first hour to the last

The course runs on the order history of Olist, a Brazilian e-commerce marketplace published as open data: about 100,000 orders from 2016 to 2018, across nine linked tables. Each group receives a management brief in the first session and defends its recommendation in a pitch at the end.

  • One case, all the way through. One analysis grows from session to session, instead of a new toy example every week.
  • See it, then do it. Each technique is first shown as a complete worked example, then practised with support, then applied by the group to its own brief.
  • Prepare, practise, apply. A short preparation before each session (one to two hours) frees class time for practice rather than lecturing.
  • Feedback early, grades late. The three milestones are small and come back with feedback, so that problems surface before the final submission rather than in it.
Table 5: Student workload.
TypeHours
Lectures and guided demonstrations8
Applied workshops and group project10
Overall teaching18
Preparation before the sessions (readings and warm-ups)7
Milestone 14
Winter-break project work, including Milestone 212
Final brief, slides and AI-use appendix8
Pitch rehearsal2
Revision for the written exam3
Personal work36
Overall student workload54

Tips for doing well

  • Start from the decision, not from the data. A beautiful dashboard that answers no question earns little.
  • Write down every cleaning choice as you make it. You will need it for the brief, and for the questions after your pitch.
  • Check AI outputs against the data, every time. Errors that look plausible are the hardest to spot.
  • Rehearse the pitch out loud, with a timer, at least twice.

6. Tools and data

Everything runs in a web browser and is free. You need a laptop in every session; for the online session, a laptop works much better than a phone.

Table 6: Tools.
ToolUsed for
Google Sheets or Excel for the webBrowser, with a free Google or Microsoft accountCleaning, pivot tables
Data Studio, formerly Looker Studio (free version)Browser, with a Google accountDashboards
DuckDB in the browsershell.duckdb.org; ready-to-query copies of the case tables are providedSQL
Google ColabBrowser, with a Google account; prepared notebook, no code to writeGuided predictive models
An AI assistant of your choiceChatGPT, Claude, Copilot, Gemini or Le Chat; free tiers are enoughSession 5 workshop
Olist Brazilian E-Commerce Public DatasetProvided in class; licence CC BY-NC-SA 4.0The running case

Data Studio and Colab require a Google account. If you prefer not to use your personal account, create a separate one for the course. Orange Data Mining, a free desktop program with a drag-and-drop interface, is a good option to go further after Session 4; it is optional.

7. Bibliography

Essential reading

  • Provost, F. and Fawcett, T. (2013). Data Science for Business: What You Need to Know about Data Mining and Data-Analytic Thinking. O'Reilly Media. Chapters 1 and 2 before Session 1; chapters 5 and 7 before Session 4.
  • Knaflic, C. N. (2015). Storytelling with Data: A Data Visualization Guide for Business Professionals. Wiley. Chapters 1 to 3 before Session 2.
  • Davenport, T. H. and Harris, J. G. (2017). Competing on Analytics: The New Science of Winning. Updated edition. Harvard Business Review Press. Background for the whole course; no chapters are assigned.

Supplemental reading

  • Berinato, S. (2019). Data Science and the Art of Persuasion. Harvard Business Review, 97(1), 126–137.
  • Perkins, M., Roe, J. and Furze, L. (2025). Reimagining the Artificial Intelligence Assessment Scale: a refined framework for educational assessment. Journal of University Teaching and Learning Practice, 22(7). doi:10.53761/rrm4y757
  • Wexler, S., Shaffer, J. and Cotgreave, A. (2017). The Big Book of Dashboards, Wiley.
  • Regulation (EU) 2024/1689 on artificial intelligence (AI Act), as amended by Regulation (EU) 2026/1744; Regulation (EU) 2016/679, the General Data Protection Regulation.

Internet resources

Formatting your references. Google Scholar gives each source in BibTeX (Cite, then BibTeX); paste them into latexci.com, a free formatter I built, choose APA 7th edition and copy the list into Word or Google Docs with its italics and hanging indent. No account. Check each entry against the source before you submit.

8. Course policies

  • Attendance. With six sessions, each one carries real weight. Attendance is expected at every session and required for Session 6; the attendance sheet is signed every morning and afternoon. If you cannot attend, tell me and the programme coordinator beforehand.
  • Absence from an assessed activity. Under the programme rules, an absence from an assessed activity (the written exam, the pitch, Milestone 3 written in class) is graded 0, and its weight is not transferred to the other assessments. A documented reason is handled with the programme coordinator.
  • Online session. Please keep your camera on during group work. The session is interactive and is not recorded.
  • Groups. Groups are fixed in Session 1. If your group runs into difficulties, come to me early: problems are much easier to solve in December than in January.
  • Accessibility. If you have an accommodation plan, for example extra time in written exams, let me know before 18 December 2026 so that it is in place for the exam.
  • Academic integrity. University rules apply to all assessed work. The AI policy above explains what is allowed for each task.
  • Questions. Email me at any time at [email protected]; I reply within two working days. I stay for 15 minutes after each on-campus session, and online meetings can be arranged on request.

Syllabus version 1.2, 1 November 2026. Programme coordinator: Véronique Flambard-Vigeant.

Instructor

Sitraka Forler, systems architect and data scientist working in industry. The course is built so that every tool runs in a web browser, for free.

Work with me →About →