Tools for business learning
Start with a business question, then choose a tool whose behaviour you can inspect. These three introductory guides connect technical practice to records, decisions and responsibility. They are original teaching, not vendor endorsements or a complete ecosystem catalogue. Official learning pages were checked on 3 October 2026; hosted services, certification and support may have separate terms.
Python: repeatable calculations and validation
Python is a programming language. In this library, it turns a written decision rule into an inspectable calculation. A finance learner can reproduce a contribution calculation; an operations learner can enumerate feasible plans; an analyst can check records before aggregating them. The business value comes from a correct, maintained process, not the mere presence of code. Begin with variables, arithmetic, comparisons, loops and functions before introducing large libraries or AI-generated programs.
The official tutorial covers control flow, data structures, functions, modules, exceptions and files. It assumes basic programming understanding; it is not designed as a complete first course for someone entirely new to programming. The interpreter and standard library are freely available. Follow the library's algorithms chapter first if those prerequisites are unfamiliar.
Our practice framework is predict, implement, verify and explain. Use the product-mix script, predict its base result, then change one capacity and explain the consequence. Keep the data and business assumptions beside the code. Test invalid inputs and preserve useful errors. An employee should understand why a result changed before a manager uses it for planning. Running a script successfully proves neither that the data is accurate nor that the recommendation fits the organisation.
Official free Python tutorial · Programming foundations · Runnable product-mix practice
Checkpoint: what remains a human decision?
The objective, missing constraints, acceptable impacts and authority to act. Code calculates consequences within the specified model.
SQL and PostgreSQL: shared records with explicit meaning
SQL expresses operations on relational data; PostgreSQL is a database system in which learners can practise those operations. A business needs stable identifiers and clear relationships between customers, orders and payments. Linking these records supports reconciliation and departmental coordination, but a join can also multiply rows when several matches exist. Understand the intended record grain before calculating totals. A correct query can still answer the wrong question.
The official PostgreSQL tutorial introduces relational concepts and SQL through hands-on work. Its progression includes tables, queries, joins, aggregates and updates. General computer familiarity is assumed; particular Unix or programming experience is not required. The documentation is publicly readable. The tutorial is introductory rather than comprehensive, and operating a real installation requires further administration study.
Use a define, query, reconcile and explain framework. Create synthetic customers and orders in a new practice database, count orders before joining, and check whether the joined result preserves the intended unit. Compare COUNT(*) with the number of distinct order identifiers. IT implements permissions and constraints; business owners define valid statuses and corrections. Do not practise updates on an organisational production database. The supplied database.sql starter is SQLite-compatible: it is a related relational exercise, not a tested PostgreSQL port. Adapt dialect and setup deliberately.
Official free PostgreSQL tutorial · Relational foundations · SQLite-compatible starter
Checkpoint: why can a join inflate a total?
One order can match several related rows. Summing its value after that join counts it repeatedly unless the query deliberately accounts for the relationship.
Git: reviewable changes and collaboration
Git supports version control: a learner can record changes to code and explanatory files, inspect differences and return to an earlier recorded state. It helps make a project's reasoning reviewable. A team should be able to tell which calculation changed, why it changed and which evidence justified the revision. Git does not establish that the explanation is truthful or that a change is correct; review and testing still matter.
The Pro Git book is freely readable on the official Git website. Use its introductory and basics material before branching and collaboration topics. Git and a hosted code-sharing service are different parts of a workflow. Do not assume a hosting account is necessary for the local practice exercise, or that version history automatically provides a complete backup strategy.
Practise in a new folder containing only synthetic teaching files. Record the initial product-mix script, change one explanation, inspect the difference and record the reason. Compare the current version with the previous one. For collaborative work, agree who proposes, reviews and approves changes. A conflict needs a considered choice about meaning; automatically selecting one side can discard valid work. Keep passwords, private customer data and confidential material outside the exercise history.
Free Pro Git book · Software engineering foundations
Checkpoint: does a recorded change prove quality?
No. It provides traceable history. Quality depends on the requirements, reasoning, tests and review applied to that change.
One integrated portfolio
Define a business decision, model synthetic records, calculate a result and preserve the reviewed changes. Submit a data dictionary, prediction, output, exception case, version explanation and managerial memo. Ask who owns each assumption, whose record is authoritative and how an incorrect result would be corrected. This combines computing practice with business judgement rather than treating tools as isolated badges.