Foundations, optional code

Data science course for professionals: think like a data scientist, with optional code

This data science course teaches working professionals to turn business questions into data questions, read descriptive statistics and charts, judge experiments and A/B tests, work well with analysts and use AI to explore data. It starts without code, with optional Code lessons in Python and SQL for hands-on analysis and machine learning. It is not a full-time bootcamp or a degree, and it will not prepare you for a data scientist job on its own.

For you or your whole team

Start with a useful first step.

7-day free trial. No credit card needed.

You will continue on the Tosheo.ai app to create your account and review the trial terms.

What you will learn

The modules in this course.

Framing data questions

Turn a vague business question into one data can answer.

Judging data quality

Spot samples, gaps and definitions that make data misleading.

Reading descriptive statistics

Use averages, medians, spread and rates correctly.

Reading and choosing charts

Pick charts that answer the question and catch misleading ones.

Judging experiments and A/B tests

Recognise a fair test and why correlation is not causation.

Briefing analysts and using AI

Write clear analysis briefs and explore data with AI while checking it.

Example task / fictional source material

Capstone: critique a claim and brief a better analysis

You receive a one-page summary from a fictional retailer claiming a loyalty scheme raised average spend, with a small table and a chart. Write a critique of up to 300 words, then a brief for an analyst describing the question, the data needed, the comparison group and the measure that would settle it.

Review criterion

Flaws identified

Names at least two specific weaknesses, such as no comparison group or a misleading axis.

Review criterion

Correct statistics

Uses averages, medians or rates correctly when discussing the table.

Review criterion

Sound design

The brief defines a fair comparison and a success measure chosen in advance.

Review criterion

Clear for an analyst

An analyst could start work from the brief without a follow-up meeting.

Copy and adapt

Prompts to try

Replace the parts in square brackets with your own details, and check every output before you use it.

Stress-test a conclusion

Here is a conclusion from a report: [paste conclusion and summary numbers from a public or made-up dataset]. List the alternative explanations that could produce the same numbers, and for each, what extra data would rule it in or out.

Explain a statistic

Explain what [statistic, for example a median or a confidence interval] means in this sentence, for a manager with no statistics training, in under 100 words: [paste sentence]. Then say what it does not tell us.

Draft an A/B test plan

Draft a simple A/B test plan to check whether [change] improves [measure] for [type of customer]. Include the control group, the one primary measure, what would count as a meaningful difference and the risks of stopping early. Mark any assumptions clearly.

Every lesson

20 lessons you can start today.

Your diagnostic picks the order and skips what you already know. Start any lesson with a 7-day free trial, no credit card needed.

What data science is forExplain what data science answers and how a project runs
FoundationStart this lesson
Asking a good data questionTurn a business question into one you can answer with data
FoundationStart this lesson
Cleaning data with AI helpFind and fix duplicates, gaps and inconsistent values
FoundationStart this lesson
Describing data: averages and spreadUse mean, median, range and outliers correctly
FoundationStart this lesson
Charts that tell the truthChoose the right chart and avoid misleading ones
FoundationStart this lesson
Exploring a dataset with AIAsk an assistant to summarise and find patterns, then check them
PractitionerStart this lesson
Pivot tables and summariesSummarise data by group in a spreadsheet
FoundationStart this lesson
Correlation is not causationExplain why two numbers moving together proves little
FoundationRead the free lesson Start this lesson
Experiments and A/B testsRead a test result and judge whether it is reliable
PractitionerStart this lesson
Sampling and how much data is enoughExplain sample size and bias in plain words
PractitionerStart this lesson
Dashboards and KPIsChoose a few measures that drive decisions
PractitionerRead the free lesson Start this lesson
Telling the story of your numbersWrite a one-page data summary for a manager
FoundationStart this lesson
Working with analysts and data teamsBrief an analyst and question their findings
PractitionerStart this lesson
Data privacy in analysisRemove personal data before analysis and sharing
FoundationStart this lesson
Your first Python data notebookLoad, inspect and chart a dataset in a notebook
PractitionerCodeStart this lesson
Data cleaning with pandasFix types, gaps and duplicates in code
PractitionerCodeRead the free lesson Start this lesson
SQL for analysisQuery, join and group tables
PractitionerCodeStart this lesson
Querying data with SQL and AIWrite and check SQL with an AI assistant
PractitionerCodeStart this lesson
Visualisation in PythonBuild clear charts with a plotting library
PractitionerCodeStart this lesson
Reproducible analysisShare a notebook others can rerun and check
AdvancedCodeStart this lesson

Who is this data science course for?

It is for managers, marketers, product people, operations leads and founders who receive dashboards, commission analysis or make decisions from data, and who want to ask better questions and spot weak conclusions. You need basic spreadsheet familiarity and nothing more.

Who is this data science course not for?

It is not for people who want a full-time route into a data scientist job. The core path needs no code, and the optional Code lessons in Python and SQL start from zero, but they do not replace the depth, project portfolio and mentoring of a university or college data science programme or an intensive full-time bootcamp, and there is no job guarantee. It is also not for people who already work as data scientists or engineers and want advanced specialist training. Many people combine the two: foundations and first code here, and a longer technical route for job-ready skills.

What will I learn, module by module?

The six modules follow the life of a real analysis, from the question someone asks to the decision they make.

1. From business question to data question

Turning 'why are sales down?' into something data can answer, and deciding what would change your mind.

2. Data: where it comes from and how it misleads

Samples, missing data, definitions that differ between teams and why the most available data is often the wrong data.

3. Descriptive statistics

Averages, medians, spread, percentages and rates, and when each one hides the story.

4. Visualisation

Choosing a chart that answers the question, and reading charts that are accidentally or deliberately misleading.

5. Experiments and A/B tests

Control groups, sample size, significance in plain language and why correlation is not causation.

6. Working with analysts and AI

Writing an analysis brief, reviewing results and using AI assistants to explore data while checking their arithmetic.

How does the personalised path adapt to me?

The diagnostic checks how comfortable you are with percentages, averages and charts, and what decisions you make. A marketing lead's examples centre on campaign tests and conversion rates; an operations manager's on delivery times and defect rates. If you already read statistics confidently, the path moves quickly to experiments and briefs.

What do I practise?

You practise the decisions that sit around data work, using made-up or public datasets. Each task is marked against a visible rubric, with feedback on what to strengthen, and you can flag any score for review.

Example task: a fictional meal-kit company, Greenbasket, ran a new welcome email for two weeks and reports that customers who opened it were 30 percent more likely to order again. You explain why this does not prove the email worked and design a fair A/B test. The rubric checks that you name the self-selection problem, define a control group, pick one success measure in advance and say how long the test should run before anyone looks at results.

How long does the data science course take?

Lessons are 10 to 15 minutes each. Your total path depends on the diagnostic and how often you study, so there is no fixed hour count. Someone with a statistics background may skip much of module three, while someone new to numbers gets extra practice before experiments.

What do I get when I finish?

A final practical assessment asks you to critique an analysis and write a brief for a better one. Pass it and you earn a recognised certificate in data science foundations for professionals. It is not an accredited qualification and does not claim technical data science competence.

How much does the data science course cost?

There is no separate course fee: it comes with your plan. Individual is USD 79 a month or USD 790 a year. Team is USD 189 a month or USD 1,890 a year for up to 3 members, then USD 49 a month or USD 490 a year per extra member, with volume pricing from 25 members. A 7-day free trial is included, and no credit card is needed.

Before you decide

A few good questions.

Does this data science course teach Python or coding?

Yes, optionally. You start without code; Code lessons in Python and SQL are optional and start from zero. For job-ready depth in model building, a college programme or a full-time bootcamp is a better fit.

Can this data science course get me a job as a data scientist?

Not on its own. It builds the judgement to work with data and analysts, and the optional Code lessons give you a first hands-on start. Data scientist roles usually need deeper programming, statistics and project experience than this course covers, and there is no job guarantee.

Do I need to be good at maths?

No. You need everyday comfort with percentages and averages. The course explains statistics in plain language with worked examples from your role.

How is AI used in this course?

You practise asking AI assistants to summarise and explore made-up or public data, then checking their arithmetic and conclusions, because AI tools can make calculation errors.

Is the certificate accredited?

No. You earn a recognised certificate after a final practical assessment. It shows applied foundations, not an accredited degree or technical competence.

Give your team a path, not another playlist.

Explore role-specific lessons, practical tasks and the manager view.

Start free trial7-day free trial · No credit card needed

Data science course for professionals: think like a data scientist, with optional code

Start with a useful first step.

The modules in this course.

Framing data questions

Judging data quality

Reading descriptive statistics

Reading and choosing charts

Judging experiments and A/B tests

Briefing analysts and using AI

Capstone: critique a claim and brief a better analysis

Flaws identified

Correct statistics

Sound design

Clear for an analyst

Prompts to try

Stress-test a conclusion

Explain a statistic

Draft an A/B test plan

20 lessons you can start today.

Who is this data science course for?

Who is this data science course not for?

What will I learn, module by module?

1. From business question to data question

2. Data: where it comes from and how it misleads

3. Descriptive statistics

4. Visualisation

5. Experiments and A/B tests

6. Working with analysts and AI

How does the personalised path adapt to me?

What do I practise?

How long does the data science course take?

What do I get when I finish?

How much does the data science course cost?

A few good questions.

Keep exploring.

Guide to AI for data analysis

Compare Individual and Team pricing

See all courses

Hands-on spreadsheets: the data analytics course

Machine learning concepts for professionals

Give your team a path, not another playlist.