Data, Analytics and AI
Data analysis fundamentals
Before a query or a chart there is a question, and most weak analysis fails there rather than in the arithmetic. This page follows the analyst's workflow from framing a request to handing over a result somebody can act on, sets out the vocabulary that carries between every tool, and names the habit that separates a junior who reports numbers from one who knows what they are entitled to say.
Why this topic exists: Ask, prepare, process, analyse, share: the analyst's workflow as Google's programme and DP-900's core data concepts teach it, before any tool (SFIA Data analytics).
Data analysis is answering a specific question with records and stating how far the answer can be trusted. The glossary entry makes that second half its centre, and rightly, because a figure delivered without its limits will be quoted without them. Most of the craft is independent of software. SFIA describes the analytics skill as getting insight out of data, tidy or messy, and communicating it so that decisions follow. Microsoft's DP-900 opens its first skill area by asking candidates to tell structured, semi-structured and unstructured data apart, and transactional workloads from analytical ones. Those distinctions decide where data lives and how you may query it, long before anyone argues about tools.
Six steps, in the order they go wrong
- Ask. Turn a request into a question with a decision attached. How are sales doing has no answer; did the spring discount raise repeat orders among existing customers has one, and it tells you which data to fetch.
- Prepare. Find the sources, learn who produced them and for what purpose, and check that they cover the period and population the question needs.
- Process. Clean and reshape until the table can bear the question. This is data wrangling, and every decision taken during it belongs somewhere it can be run again.
- Analyse. Compare, aggregate, segment. Mostly this is SQL or a spreadsheet, and the discipline is to set each figure against a baseline instead of admiring it alone.
- Share. State the finding in one sentence, then show the chart that supports it, not the other way round.
- Act. Agree what the decision-maker will do with the result, and which result would have changed their mind. Without this step the analysis is merely a report.
Google's Data Analytics programme devotes one course to each of those steps, the wisest choice in its design: beginners who learn tools one at a time never see how a careless second step poisons the fifth. Every other framework on this site hides the same sequence under different names, from the modelling stages of machine learning to the preparation area of the Power BI exam.
Words that carry between tools
A handful of terms move an analyst between Excel, SQL, Python and any BI product. Granularity is what one row represents, whether a customer, an order or an order line, and most wrong totals come from joining two tables held at different grains. A dimension is something you group by, a measure something you add up. A metric is a measure with an agreed definition and an owner. A baseline is what you compare against, and a segment is a slice of the population chosen before you look at the results, never after. Learn these and the tools become interchangeable.
The beginner's mistake
Reporting a difference because it exists. Two numbers drawn from the same imperfect source will almost never match, and the real question is whether the gap exceeds the noise, the data errors and the seasonal wobble combined. Juniors present every movement; experienced analysts say which ones they would not act on. The natural next pages are SQL for analysis and statistics and experimentation, and IBM's free Data Fundamentals badge or Google's Data Analytics certificate test the ground covered here.
Next to this topic
- SQL for analysisQuerying, joining, aggregating and windowing data is the analyst's first real tool and the language of every warehouse (DP-900 relational concepts; DMBOK Data Storage and Operations; the programming field owns database design).
- Data visualisation and business intelligenceDashboards, reports and the semantic models behind them: PL-300's four skill areas and DMBOK's Data Warehousing and BI area (Tableau, Power BI, Looker).
- Statistics and experimentationProbability, inference, regression and A/B tests separate a chart from a claim (Google's Advanced Data Analytics; SFIA Data science).
- Python for data workpandas, NumPy, notebooks and plotting: the working environment of analysis and machine learning (SFIA Data science; the AI pillar's argument for Python).
- Data engineering and pipelinesIngestion, transformation, warehouses, lakehouses and orchestration (DMBOK Data Integration and Interoperability; DEA-C01; DP-700; Databricks; Spark, Airflow, dbt).
- Data governance, quality and privacyOwnership, definitions, metadata, master data, quality rules and lawful use are DMBOK's centre and half its wheel; nobody hires for them until the numbers disagree.
- Machine learningSupervised and unsupervised learning, model evaluation, deep learning and the discipline of not fooling yourself (SFIA Machine learning; AIF-C01; Azure AI Fundamentals).
- Generative AI, LLMs and agentsFoundation models, prompting, retrieval, fine-tuning and agents: the newest layer of the field and the one AIF-C01 and Azure AI Fundamentals now test (the AI portal reads the concepts in order).
- Natural language processing and computer visionText and images are the two applied fields where classic machine learning met deep learning first, and where most production AI still lives (SFIA Machine learning; Azure AI Fundamentals AI-901 workloads).
- Responsible AI and AI governanceBias, transparency, privacy, safety and the frameworks that make them work (NIST AI RMF; AIF-C01's responsible-AI domain; SFIA 9's AI and data ethics skill; the EU AI Act).
- MLOps and AI engineeringGetting a model or an LLM application into production and keeping it there: deployment, monitoring, evaluation, cost (SFIA Machine learning and Data engineering; Applied Skills in Foundry; MLOps in the AI portal).
Concepts to know
Glossary entries with the reason each one matters here.
- Data Analysis
The topic's own concept.
- Data Wrangling
Most analysis time is spent here.
- SQL
The first tool after the spreadsheet.
- Data Visualization
How findings are shown.
Certifications that test it
Vendor exams and free certificates; facts, cost and the preparation path are on each page, and the certifications hub has them all.
- Google · Course completionGoogle Data Analytics Professional CertificateGoogle's programme is this workflow.
- IBM · Course completion · freeData FundamentalsIBM's free introduction.
- Microsoft · DP-900Microsoft Certified: Azure Data FundamentalsCore data concepts, Microsoft's version.
- freeCodeCamp · Graded projects · freeData Analysis with Python CertificationfreeCodeCamp's free project-based certificate.
Tools of the trade
- Microsoft Excel
Where most analysis still happens.
Frequently asked
- Do I need Excel before I learn SQL?
- You need what a spreadsheet teaches: sorting, filtering, a pivot table and a feeling for when a total looks wrong. If work has already given you that, go straight to SQL. If not, a few weeks in a spreadsheet makes the ideas visible before you meet them as code.
- What separates data analysis from business intelligence?
- Analysis answers a question, often once. Business intelligence builds the models and reports that answer recurring questions for many people with no analyst in the loop. Analysts usually build the first version of a metric and BI makes it permanent; see data visualisation and BI.
- How can I practise without a job in data?
- Pick a public dataset about something you understand, write down the question and the decision it would inform before opening the file, and publish the whole chain: question, cleaning steps, result and what you would refuse to claim. One such write-up tells an employer more than a stack of completed tutorials.
Courses in the directory
1,806 courses are filed here; the top 6 by our ranking, details and the provider link on each course page.
Last reviewed 26 September 2026 · Getting Digital