Data, Analytics and AI
Data visualisation and business intelligence
A chart answers one question for one reader; business intelligence answers the same questions every morning for hundreds of readers who never meet the analyst. This page is about that second job: the semantic model under a dashboard, the measures everyone has to agree on, the refresh and permission settings that keep it honest, and why Power BI, Tableau and Looker differ less than their users like to argue.
Why this topic exists: Dashboards, 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).
Data visualisation is the craft of making a comparison visible. Business intelligence is the organisation built around it: the feed into a reporting model, the model that fixes what revenue or an active customer means, the reports on top, and the rules about who may see which rows. DMBOK files it together with data warehousing as a single knowledge area, while SFIA splits it into a business intelligence skill for recurring management information and a separate skill for the graphical side. The split shows up in hiring. A BI developer spends more of the week on models and refresh schedules than on colour palettes.
What the PL-300 blueprint actually asks
Microsoft's Power BI Data Analyst exam is the most detailed public description of BI work there is, and it rewards reading even for people who will never open Power BI. Its four skill areas correspond to tasks that exist in any BI stack, and each has a characteristic place where newcomers come unstuck.
| PL-300 skill area | The work in any tool | Where people get stuck |
|---|---|---|
| Preparing data | Connecting to sources, profiling columns, fixing nulls and types, shaping fact and dimension tables | Cleaning inside every report instead of once, upstream |
| Modelling | Relationships, a shared date table, measures written once and reused everywhere | Many-to-many relationships and filters that flow the wrong way |
| Visualising and analysing | Choosing visuals, drill paths, accessibility, spotting outliers and trends | Pages that show everything and say nothing |
| Managing and securing | Workspaces, scheduled refresh, gateways, row-level security, sensitivity labels | Sharing by export to spreadsheets, which leaves no trail |
The model is the product
Beginners judge a BI tool by its charts. Practitioners judge it by its modelling layer, whether that means DAX measures in Power BI, calculated fields and published data sources in Tableau, or LookML in Looker. All three rest on the same star schema underneath: narrow fact tables of events, wide dimension tables describing them, and measures defined once so that every page adds up the same way. Get that right and the visuals nearly draw themselves; get it wrong and every report needs a footnote. The other decision that shapes the experience is whether data is imported into the model or queried live at the source, the difference between a dashboard that feels instant and one that stalls on every click.
The classic mistake is building a new dashboard for every request. A year later the organisation has hundreds of them, three definitions of margin and nobody who knows which is current. Sound BI teams certify a small number of shared models and let colleagues build their own pages on top. PL-300 examines the Microsoft version of this discipline; Google's Business Intelligence certificate covers similar ground more briefly and without an exam. Where the agreed definitions ought to come from is the subject of governance, quality and privacy.
Next to this topic
- 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).
- 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 Visualization
The topic's own concept.
- Data Analysis
A dashboard is an analysis that keeps running.
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.
Tools of the trade
- Microsoft Excel
Charts and dashboards before BI tools.
- Microsoft Power BI
The BI tool PL-300 certifies.
- Tableau
The visualisation tool analysts learned first.
Frequently asked
- What is a semantic model?
- The layer between raw tables and reports that holds relationships, measures and business-friendly names, so that everyone who asks for margin gets the same calculation. Power BI uses the term directly; other tools speak of a data model, an explore or a published source. It is the part of BI work that outlives any single dashboard.
- How much DAX does PL-300 expect?
- A working level: aggregations, CALCULATE and the filter context it changes, time intelligence and semi-additive measures such as account balances. The exam also expects Power Query for preparation. Neither needs a programming background, but both need practice on a model with more than one fact table.
- Is a dashboard the same thing as a report?
- Not quite. In Power BI a report is a set of pages built on one model for exploration, while a dashboard pins the most important tiles from several reports onto a single screen for monitoring. More generally, a dashboard tells you whether things are normal and a report helps you find out why they are not.
Courses in the directory
411 courses are filed here; the top 5 by our ranking, details and the provider link on each course page.
Last reviewed 26 September 2026 · Getting Digital