Data, Analytics and AI
Data work runs from a question somebody asks on a Monday morning to a model answering thousands of questions nobody reads one by one, and the people along that line rarely share a job title. This hub divides the field the way its reference frameworks do, by the kind of work and the knowledge behind it, then shows which of twelve topics each role lives in and which credentials say anything about them.
The field this hub covers begins before any software is opened. Somebody wants to know why renewals slipped, whether a price change worked or which invoices look wrong, and the work is to turn that wish into a question records can settle, find those records, repair them, compute something that survives scrutiny and put it in front of the person who asked. Everything beyond that is the same activity at larger scale or with less human attention: pipelines so the records arrive by themselves, models so an answer is produced for every new case, and generative systems that write rather than count. Our data and analytics pillar argues which stage of that line hires first; this page is the map underneath the argument.
Two frameworks, twelve topics, seven jobs
Two references divide the territory without selling anything. DAMA's Data Management Body of Knowledge draws a wheel of eleven knowledge areas: governance at its hub, the other ten disciplines round the rim, from architecture and modelling to metadata and quality. It is a data manager's view and says little about models. SFIA 9, the skills framework many employers borrow when they write role profiles, fills that gap: its data and analytics group separates management, modelling and design, database design, analytics, data science, machine learning, business intelligence, engineering and visualisation, and a neighbouring skill covers the ethics of AI and data. The table places each topic against the nearer of the two frameworks and against the role whose working week it mostly fills. The placements are our reading, and real adverts blur several rows into one post.
| Topic | Nearest DMBOK area or SFIA skill | Who does it for a living |
|---|---|---|
| Data analysis fundamentals | SFIA data analytics | Data analyst |
| SQL for analysis | DMBOK Data Storage and Operations | Analyst, analytics engineer |
| Data visualisation and BI | DMBOK Data Warehousing and BI; SFIA business intelligence | BI developer |
| Statistics and experimentation | SFIA data science | Data scientist, product analyst |
| Python for data | SFIA data science | Data scientist, analyst |
| Data engineering | DMBOK Data Integration and Interoperability; SFIA data engineering | Data engineer |
| Data governance, quality and privacy | DMBOK Data Governance, Metadata, Data Quality, Reference and Master Data | Data steward |
| Machine learning | SFIA machine learning | Data scientist, ML engineer |
| Generative AI, LLMs and agents | SFIA machine learning | AI engineer |
| NLP and computer vision | SFIA machine learning | ML engineer |
| Responsible AI | SFIA AI and data ethics | AI governance lead, data steward |
| MLOps and AI engineering | SFIA machine learning and data engineering | ML engineer, AI engineer |
Credentials, and the doors in
Credentials here come in three kinds, and each proves something narrower than its marketing. The gentle starts are courses with a badge at the end: IBM's Data Fundamentals and AI Fundamentals cost nothing, Google's Data Analytics certificate runs on a monthly subscription with no final exam, and all three show that you finished a structured syllabus. Vendor fundamentals papers such as DP-900, AWS AI Practitioner, Azure AI Fundamentals and Oracle's free OCI AI Foundations prove vocabulary on one platform under exam conditions. Associate papers prove you can build something there: PL-300 a Power BI model and report, DEA-C01, DP-700 and the Databricks associate a pipeline. Not one of them demonstrates that you picked the right question, or that you can hold a figure steady when the person who commissioned it hoped for another, and that is what analyst interviews spend their time probing.
- Spreadsheet users already do half of data analysis fundamentals. Add SQL next, because it removes the copy-and-paste ceiling that limits what a workbook can handle, then one BI tool. Python can wait until a problem demands it.
- Developers tend to underrate how much of the field is definitions rather than code. The quickest route in is data engineering, where version control and testing are scarce habits, or Python for data followed by machine learning if the mathematics appeals.
- Developers heading for AI products can begin at generative AI and MLOps, provided they learn evaluation early, since a language-model feature without test cases is a demo. The AI portal reads the thirteen AI concepts in a sensible order.
- Compliance, risk and operations staff start at governance and responsible AI, where their instinct for ownership and evidence is the missing ingredient.
Governance, the centre nobody staffs
DMBOK puts governance in the middle of its wheel because every other area assumes it. A quality rule needs someone who owns the definition it tests, a dashboard needs an agreed meaning for each measure, and a model needs a record of what it learned from and whether that use was lawful. In most organisations none of this has a budget line until two reports reach the board with different revenue totals. Then a data steward is appointed, a glossary appears and a quarter disappears into reconciling figures that could have agreed from the outset. The same pattern repeats one level up with AI, where accountability and recourse arrive after the pilot instead of before it; the AI pillar explains why that order stalls projects. People fluent in both the technical and the governance language are uncommon, and the frameworks above are a fair place to acquire the second.
The field, part by part
Each topic says why it exists and what it rests on; the order is the order a newcomer would take.
- Data analysis fundamentalsAsk, 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).
- 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 of this field.
- Agentic AI
Agentic AI is software that wraps a language model in a loop where it can request actions from real tools, read what each action returns, and choose the next move until the job is finished or a rule stops it.
- CV
Computer vision is the branch of artificial intelligence that turns images and video into structured answers: what is in the frame, where it sits, which pixels belong to it, and where it moves next.
- Data Analysis
Data analysis is the practice of interrogating data to answer a specific question, and of establishing how much weight the answer can bear.
- Data Engineering
Data engineering is the discipline of moving records from the systems that produce them into a store where they can be queried, and of keeping that supply correct while the sources underneath keep changing.
- Data Visualization
Data visualisation is the practice of encoding numbers as position, length and colour so that a comparison a reader would otherwise have to calculate becomes something they can simply see.
- Data Wrangling
Data wrangling is the work of repairing and reshaping real records, and of deciding what their gaps and inconsistencies mean, until a table can be trusted with the question being asked of it.
- Deep Learning
Deep learning is machine learning built on neural networks of many stacked layers, which work out for themselves which properties of the raw data matter instead of being handed a list by a person.
- GenAI
Generative AI is the market's name for models whose output is a piece of content, such as text, images, audio, video or code, as opposed to the much larger body of models that sort, score or forecast things that already exist.
- LLMs
Large language models are neural networks trained on a vast quantity of written text to predict the piece of text that comes next, an objective narrow enough to state in a line and broad enough to yield writing, translation, code and summary as side effects.
- ML
Machine learning is the branch of computing in which a program works out its rule from examples instead of being handed that rule by a programmer, and is judged on how well the rule holds on cases it was never shown.
- MLOps
MLOps is the engineering practice of keeping a trained model useful in production: recording what produced it, serving it, watching for the day its answers stop matching the world, and retraining or reverting before anyone downstream is harmed by the gap.
- NLP
Natural language processing is the area of AI that makes human language readable and writable by machine: sorting text into categories, pulling structured facts out of it, translating it, generating it and measuring whether any of that came out right.
- Neural nets
A neural network is a stack of layers of very simple arithmetic units whose connection strengths are adjusted by training until the whole arrangement turns the inputs you have into the outputs you want.
- Prompt Engineering
Prompt engineering is the practice of composing what you send a generative model, the instruction, the supporting material and the worked examples, so that it returns something you can actually use.
- Responsible AI
Responsible AI is the practice of building and running systems whose decisions you could defend to the person on the receiving end of them, covering fairness, transparency, privacy and accountability.
- RAG
Retrieval-augmented generation is a pattern that searches a body of documents when a question arrives and places the passages it finds into the model's prompt, so the answer is drawn from those sources rather than from training alone.
- SQL
SQL is the declarative language for asking questions of relational databases: you state which rows and columns you want and how tables relate, and the database engine works out how to retrieve them.
- Transformer
A transformer is a neural-network design that takes a whole sequence in at once and, for every element in it, works out how strongly each of the other elements should influence that one.
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.
- AWS · AIF-C01AWS Certified AI Practitioner
- AWS · DEA-C01AWS Certified Data Engineer – Associate
- Databricks · Data Engineer AssociateDatabricks Certified Data Engineer Associate
- Google · Course completionGoogle Advanced Data Analytics Professional Certificate
- Google · Course completionGoogle Business Intelligence Professional Certificate
- Google · Course completionGoogle Data Analytics Professional Certificate
- Microsoft · AI-901Microsoft Certified: Azure AI Fundamentals
- Microsoft · DP-900Microsoft Certified: Azure Data Fundamentals
- Microsoft · DP-700Microsoft Certified: Fabric Data Engineer Associate
- Microsoft · PL-300Microsoft Certified: Power BI Data Analyst Associate
- IBM · Course completion · freeAI Fundamentals: Foundations for Understanding AI
- freeCodeCamp · Graded projects · freeData Analysis with Python Certification
- IBM · Course completion · freeData Fundamentals
- Microsoft · Lab-based assessment · freeMicrosoft Applied Skills: Get started developing agents in Microsoft Foundry
- Oracle · 1Z0-1122-26 · freeOracle Cloud Infrastructure AI Foundations Associate
Neighbouring fields
- Finance and Accounting
Financial analysis starts in a spreadsheet and ends in a data pipeline; the data field owns the methods, this hub owns what the numbers mean.
- Workplace Productivity and Office Skills
Excel is where most analysts start; the data field takes over at Power BI, SQL and statistics.
- Business, Projects and Management
Analytics and BI report the business; the data field owns the methods, this hub owns the decisions they inform.
- Digital Marketing
GA4 and attribution are analytics; the data field owns the statistics, this hub the marketing questions.
- Programming and Software Development
Python and SQL are shared ground: the programming field teaches them as languages, the data field uses them as instruments.
- Cloud Computing
Warehouses, lakes and model endpoints are cloud services; the data field decides what to build, the cloud field where it runs.
Frequently asked
- Is data science a separate career from data analysis?
- Less separate than the titles suggest. SFIA lists them as distinct skills: analytics draws out and communicates insight, while data science adds statistics and predictive modelling. In practice a data scientist is an analyst who is also trusted to build and validate models, and most began as analysts. If you are choosing, start with analysis; the step up adds to it rather than replacing it.
- How much mathematics do machine learning roles need?
- Enough to read what a model is doing: probability, some linear algebra and the idea of minimising a loss. Using libraries needs less than building them, and research roles need far more. The weakest area in most applicants is not calculus but evaluation, meaning honest test sets and knowing when a score is too good to believe.
- Which cloud should my data certificate be on?
- The one your target employers run. Concepts carry between Azure, AWS, Google Cloud and Databricks with little loss; service names do not. Job adverts in your region tell you more than any ranking, and if they show no pattern, a vendor-neutral start followed by one associate exam is a sound order.
- What does a data steward actually do?
- Looks after the meaning of a set of data on behalf of the business: agrees definitions, decides who may use it for what, sets quality rules and chases fixes when they fail. It is usually a duty added to a domain expert's job rather than a separate hire, which is why it tends to happen late or not at all. The governance topic goes further.
- Where does this hub stop and the programming hub start?
- Building and operating an application's database belongs to developers and sits under databases and SQL; storage services belong to cloud. This hub begins once the data exists and someone wants to ask it something, follows that question through to a model in production and covers the rules around it. Protecting data from attackers is a cybersecurity subject.
Courses in the directory
3,349 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