Training for a data analyst job: degree, bootcamp or self-taught

The data analyst job attracts very varied profiles: business or engineering school students, economics or statistics graduates, career changers from management control, marketing or logistics. Training routes are just as diverse, from a university master’s degree to self-directed learning with free resources. Which one to choose depends on your starting point, your budget, the time you have and the type of employer you are targeting. This article starts with what employers really expect from a junior data analyst, compares the routes in, then shows how to prove your skills on a resume.

What an employer expects from a junior data analyst

Onisep 🔗 describes the data analyst as the person who collects, processes and analyses data to inform the company’s decisions. In a junior job ad, this translates into four expectations, which you should keep in mind when assessing any programme.

1. SQL, above all. Almost every role requires you to query a database: selections, joins, aggregations, subqueries and, increasingly, window functions. It is often the subject of a test during recruitment.
2. A visualisation tool. Power BI, Tableau, Looker or an equivalent tool, to build dashboards that non-specialists can understand.
3. A grounding in statistics and an advanced spreadsheet. Descriptive statistics, basic notions of testing, mastery of a spreadsheet for daily exchanges with teams.
4. Business understanding and communication. Turning a vague question into an analysis, then presenting a result clearly, with its limits.

Python or R increasingly complement this foundation, to automate processing and carry out more advanced analyses. The level required generally remains lower than for a data scientist, whose job is compared with the data analyst’s in our article on the difference between a data analyst and a data scientist.

Any serious programme must bring you to this level, and allow you to prove it.

Master’s degrees and engineering schools

What they bring

Long programmes (master’s degrees in statistics, econometrics, business intelligence or data science, engineering and business schools with a data major) offer:

• solid fundamentals in mathematics, statistics and computing;
• a five-year degree, required by some employers, especially large groups;
• internships and often work-study, which provide first professional experience;
• a network of alumni and relationships with companies.

Below master’s level, shorter programmes such as the BUT in data science, which Onisep presents notably through its visualisation and decision-support tools track 🔗, also lead directly to analyst or research officer roles.

Who it suits

This route suits students in initial education, and career changers who can devote one or two years to returning to study, possibly through work-study. It remains the safest route to the employers most selective about degrees.

Bootcamps and short programmes

What they bring

Data analysis bootcamps and short programmes promise rapid skill development, from a few weeks to a few months. They focus on practice: SQL, a visualisation tool, Python, projects on real datasets, sometimes a project in partnership with a company.

Some prepare for a qualification listed in the national register of professional certifications, which matters for funding through the personal training account 🔗 (CPF); the list of registered certifications is kept by France compétences 🔗.

Points to watch

• The starting level: a programme of a few weeks often assumes prior ease with figures and spreadsheets.
• The actual content: check the share of SQL, statistics and projects, rather than the list of tools presented.
• The results announced: as with coding bootcamps, look at how employment figures are calculated. Our article on what recruiters think of bootcamps details the questions to ask.

Who it suits

This route particularly suits career changers who have already handled data in their job: management controllers, marketing research officers, stock managers, sales analysts. Their knowledge of a domain makes up for a shorter programme.

Learning on your own: possible, under certain conditions

The data analyst job is one of the most accessible in tech for self-taught people. Free or low-cost resources are plentiful, the tools are available in free versions, and public datasets abound, on data.gouv.fr 🔗, the Insee 🔗 website or platforms such as Kaggle 🔗.

It is possible, under three conditions.

1. A structured plan. SQL first, up to window functions; then a visualisation tool; then descriptive statistics and tests; finally Python for data manipulation. Without a plan, you pile up tutorials without progressing.
2. Real projects. Choose concrete questions on public data or on your current work, and see them through: collection, cleaning, analysis, dashboard, conclusions. Two or three well-run projects are better than ten exercises.
3. External proof. A vendor certification, such as Microsoft’s Power BI Data Analyst certification 🔗, or a recognised online certificate, such as the Google Data Analytics certificate 🔗, can reassure a recruiter about a path without a specialised degree.

The limits of this route are real: no framework, no feedback from a teacher, no network and no internship. It works best for people already in a job who can apply their new skills in their current role, and then move internally.

Proving your skills on the resume

Whatever route you choose, the resume must prove the four expectations listed at the start: SQL, visualisation, statistics, business understanding.

• Present the programme with its real content: exact title, duration, qualification or degree obtained, and what you practised.
• Describe your projects as experiences: the question asked, the data, the method, the result and whom it served.
• Highlight SQL, with a credible level. Our article on the SQL level to claim offers a grid to place yourself.
• Rewrite your former job, highlighting what involved data: tracking spreadsheets, reporting, analyses for management.
• Add a link to a portfolio or published dashboards, if your projects lend themselves to it.

Before: "Data analyst programme: SQL, Python, Power BI, Excel, statistics."

After: "Final programme project: analysis of library attendance in a metropolitan area using the local authority’s open data; cleaning in Python, star schema in PostgreSQL, Power BI dashboard presented to a jury of professionals."

The data scientist resume template shows how to present data projects by their problem, data and effect; the same logic applies to an analyst’s resume, with more room for dashboards and indicators.

A programme is best showcased on a readable page that recruitment software can extract without losing your tools or dates. The TechCV editor handles this formatting and lets you focus on describing your projects. You can create your resume as soon as your first project is finished.

Frequently asked questions

Do you need a five-year degree to become a data analyst?

Not always. Some employers require it, notably large groups, but many roles are accessible with a three-year degree, a short certifying programme or a well-built career change. Projects and SQL mastery weigh heavily in the decision.

Is Python essential for a data analyst?

It is not always required for a first job, but it is increasingly appreciated. SQL and a visualisation tool remain priorities; Python comes next, to automate and deepen analyses.

How long does it take to train?

It depends on the starting point. Someone already comfortable with figures and spreadsheets can reach a junior level in a few months of regular work; a complete beginner will need more time, and will benefit from a supervised programme.

Is a Power BI or Google certification enough to land a job?

Rarely on its own. It complements a degree, a programme or experience, and reassures on a specific skill. Combined with well-presented concrete projects, it can make the difference at the first screening.