BI self-learning resources (cross-tool)
Tool-independent foundations first, then study orders, then per-tool pointers. Only resources that were opened or confirmed to exist on 2026-10-05 are listed. Free/paid is stated, never amounts (prices go stale; check vendor pages). Per-tool notes with fuller learning sections: power-bi-learning-path and tableau-learning-path (not yet present on 2026-10-05; see also power-bi), qlik-learning-path, looker-learning-path; tool notes: looker, qlik-cloud-analytics, metabase, apache-superset, open-source-bi-comparison. Related vault notes: Looker Studio, business-intelligence, microsoft-fabric-terminology-guide, Looker MCP server.
1. Transferable foundations
(Opinion) Most BI tools can be seen as a front end to three ideas: a well-shaped model, a query language, and a visual design discipline.
Data modelling (star schema, Kimball)
- Facts (events, numbers to summarise) and dimensions (things used to filter and group); keep the grain of a fact table consistent. (Opinion) This shape tends to suit the tools below.
- Book: The Data Warehouse Toolkit, 3rd edition (2013), Ralph Kimball and Margy Ross, Wiley. Paid book; the Kimball Group page confirms title, authors and edition.
- Free and practical: Microsoft Learn, “Understand star schema and the importance for Power BI” (page updated 2026-04-01). Covers fact/dimension tables, surrogate keys, snowflake and role-playing dimensions, slowly changing dimensions. Useful beyond Power BI; it itself points to Kimball for the full theory.
SQL for analytics
- (Opinion) SQL literacy helps even in “no-code” tools: Metabase, Superset, Sigma, Looker (LookML generates SQL) and Tableau custom SQL. Focus on joins, aggregation, window functions, CTEs, date handling.
- Metabase Learn (free) has a SQL track (syntax, filtering, joins, debugging) plus database fundamentals (table relationships, data types, normalisation).
- Alex the Analyst (YouTube, free): SQL basics/advanced videos and a “data analyst bootcamp” playlist (confirmed via Class Central listings).
- Kaggle Learn (free): has a Data Visualization course among its tutorials; the SQL course was not individually confirmed in this session (see Open items).
- The old Mode SQL tutorial URL now redirects to ThoughtSpot (Mode is part of ThoughtSpot); the redirect target was not checked.
Data visualisation principles
Books (publisher and year taken from search results, not checked on publisher pages unless noted):
- Edward R. Tufte, The Visual Display of Quantitative Information (1983): described as a classic on data-ink, clarity and graphical integrity (opinion).
- Stephen Few, Show Me the Numbers (first edition 2004) and Information Dashboard Design (2006): business-oriented tables, graphs and dashboards.
- Alberto Cairo, The Truthful Art (New Riders, 2016): charts, honesty and journalism.
- Claus O. Wilke, Fundamentals of Data Visualization (O’Reilly, 2019): the full text is free online at https://clauswilke.com/dataviz/ (confirmed; CC BY-NC-ND).
- Cole Nussbaumer Knaflic, Storytelling with Data plus the workbook Let’s Practice! and Before & After (confirmed on storytellingwithdata.com/books).
DAX vs SQL vs LookML thinking
- SQL: set-based, you write the whole query; the engine has no “current filter”.
- DAX (Power BI, Fabric, Analysis Services): measures are evaluated per cell under a filter context set by visuals and slicers; row context and context transition are the hard part. Model quality decides how simple the DAX can be.
- LookML (Looker): declarative modelling language defining dimensions, measures, joins (explores) in version-controlled files; Looker generates SQL at query time, so SQL skills transfer directly.
- Tableau: calculated fields and LOD expressions over a (often flat) data source; Qlik: associative engine plus load-script (ETL) language and set analysis. These are described in the per-tool notes.
- (Opinion) Learning SQL and star schemas first makes DAX/LookML/LOD easier to read as “SQL aggregation, but context-driven or declared once”.
Semantic layers
- A semantic layer defines business metrics and relationships once (Power BI semantic model, LookML, ThoughtSpot semantic layer/models, Superset datasets with custom metrics, Metabase models/metrics) so every report gives the same answer. The Superset docs list a “semantic layer for defining custom dimensions and metrics” among its features; ThoughtSpot docs list a “semantic layer for governed insights”.
- It matters more with AI assistants: Microsoft’s Copilot docs say that without data preparation Copilot “can misinterpret the data and return generic or inaccurate results”.
Data storytelling
- Knaflic’s Storytelling with Data (above) is a commonly recommended entry point (opinion); Data Visualization Society’s publication Nightingale carries practitioner articles (see Communities).
2. Recommended study order
From zero
- Spreadsheet basics and what a table, key and aggregation are (any free tutorial; not tool-specific).
- SQL for analytics (Metabase Learn SQL track or Alex the Analyst videos), then practise on a Kaggle or sample database.
- Dimensional modelling: read the Microsoft star schema article, then sample chapters of Kimball.
- Visual design: Knaflic then Wilke (free online) or Tufte/Few.
- Pick ONE tool and finish its official free path (in the author’s view Power BI via Microsoft Learn is a low-friction start on Windows; Tableau Public for a free desktop-class tool; Metabase or Superset if you want open source).
- Weekly practice: Makeover Monday or Workout Wednesday; publish to a portfolio.
- Optional certification for the chosen tool; then add a second tool and note what transfers.
Already know Excel
- Map concepts: Power Query = transform step; PivotTable = a visual with fields; Excel formulas = calculated columns; measures are the new idea.
- Learn the star schema first, because Excel flat tables are the common failure mode in BI tools.
- Power BI is the shortest path (Microsoft Learn “Get started with Microsoft data analytics” and “Model data with Power BI” paths); learn DAX measures after relationships.
- Add SQL in parallel; then visual design and storytelling.
- PL-300 (below) is the matching certification (opinion: a natural target).
3. Per-tool pointers
All links below were opened (or found on an official page) on 2026-10-05 unless noted.
Power BI
- Microsoft Learn Power BI hub (free): https://learn.microsoft.com/en-us/training/powerplatform/power-bi/ with paths “Get started with Microsoft data analytics”, “Model data with Power BI”, “Build Power BI visuals and reports”; hub updated 2026-06-26.
- SQLBI (free parts): DAX Guide https://dax.guide/, DAX Patterns https://www.daxpatterns.com/, “Introducing DAX” video course (marked Free, 150 minutes; https://www.sqlbi.com/p/introducing-dax-video-course/, confirmed 2026-10-08), plus videos and articles at https://www.sqlbi.com/ (the earlier “550” count is not on the site and is dropped). Books (paid): The Definitive Guide to DAX (3rd ed., Microsoft Press), Optimizing DAX, DAX Patterns (2nd ed.).
- Guy in a Cube and Curbal on YouTube (see Creators).
Tableau
- Official free training at https://www.tableau.com/learn is blocked (403 to curl and WebFetch, 2026-10-08), so its content is unverified. Opened instead: Tableau Help tutorial https://help.tableau.com/current/guides/get-started-tutorial/en-us/get-started-tutorial-home.htm and the Trailhead module “Tableau Data Analyst Certification Prep Guide” (https://trailhead.salesforce.com/content/learn/modules/cert-prep-tableau-data-analyst). See tableau-learning-path.
- Tableau Public (free) for practice and a public portfolio; Makeover Monday and Workout Wednesday are Tableau-centred (below).
Qlik
- Qlik Learning https://learning.qlik.com/ : free just-in-time videos and selected courses, subscription for more, plus an Academic Program; a “Think Like a Data Scientist” data-literacy course.
Looker
- Google Cloud Skills Boost learning path “BI and Analytics with Looker” (https://www.skills.google/paths/28; WebFetch 2026-10-08: managed by Google Cloud, 10 activities, last updated about a year ago). Course names confirmed on the Looker docs page: Get Started with Looker, Build LookML Objects in Looker, Manage Data Models in Looker, Prepare Data for Looker Dashboards and Reports (“Applying Advanced LookML Concepts in Looker” was from search results only). Free vs paid for labs: not verified.
- Google Cloud blog on the Looker certification program (2020-02-20); the certifications are not on Google’s current certification page (see looker-learning-path).
ThoughtSpot
- ThoughtSpot University https://training.thoughtspot.com/ : training site; specific course titles, learning paths and free/paid status were not verified (see Open items).
- Docs https://docs.thoughtspot.com/cloud/latest/ (Liveboards, semantic layer, Spotter agents).
Sigma
- Sigma Quickstarts and the self-paced hands-on lab guide https://www.sigmacomputing.com/go/hol : free, five modules (navigate UI, build a workbook, explore and analyse, visualisations, interactive dashboards), with a free trial environment.
- Sigma help centre https://help.sigmacomputing.com/ (references Sigma Quickstarts, Sigma Academy and certification; details not confirmed). Workout Wednesday includes Sigma challenges.
Metabase
- Metabase Learn https://www.metabase.com/learn/ : free; “get up and running in 30 minutes” videos, SQL, database fundamentals, chart-selection cheat sheet, community forum.
Apache Superset
- Official docs https://superset.apache.org/user-docs/intro/ (User, Administrator and Developer guides) and the quickstart https://superset.apache.org/docs/quickstart/ . Open source (Apache project). No official video curriculum was confirmed.
4. Certifications (exact names)
| Certification | Issuer | Free or paid | Notes |
|---|---|---|---|
| Microsoft Certified: Power BI Data Analyst Associate (exam PL-300) | Microsoft | Exam paid, price varies by country; free practice assessment and exam sandbox; renewal free via online assessment every 12 months | Intermediate; 100 minutes; proctored; assesses prepare, model, visualise/analyse, manage/secure Power BI. Page updated 2026-04-21 (page metadata; corrected from 2026-04-20). |
| Salesforce Certified Tableau Data Analyst | Salesforce/Tableau | Paid exam (fee on the Trailhead prep page); free Trailhead prep module | 60 scored questions, 65% pass, proctored (Pearson VUE or online), title valid one year; Desktop, Prep, Server/Cloud. Source: Trailhead prep unit, 2026-10-08 |
| Qlik Sense Business Analyst, Data Architect, System Administrator; Qlik AI Specialist (analytics); Talend / Qlik Replicate exams (integration) | Qlik | Paid exams bought and taken on Qlik Learning (per-exam Purchase link; amounts not recorded) | 90 minutes, 50 questions; pass marks 62% (73% for AI Specialist); Credly badges. Source: learning.qlik.com exam-detail pages, 2026-10-08 |
| Looker Business Analyst; LookML Developer | Google Cloud | Fee status unknown | Announced 2020-02-20; absent from Google Cloud’s current certification page (checked 2026-10-08), so possibly retired |
| ThoughtSpot University certifications | ThoughtSpot | not verified | not verified |
5. Practice datasets and challenges
- Makeover Monday https://makeovermonday.vercel.app/ (moved from makeovermonday.co.uk; the old URL still answers 200 but points to the new site): weekly free visualisation challenge; further details not verified.
- Workout Wednesday https://www.workout-wednesday.com/ : weekly challenges to recreate a visualisation; tracks for Tableau, Power BI, Sigma and CRM Analytics; solutions and Saturday walkthroughs; run by a volunteer contributor team.
- Kaggle https://www.kaggle.com/learn : free tutorials and datasets (Data Visualization and Python/Pandas courses confirmed on the page).
- Data Visualization Society https://www.datavisualizationsociety.org/ : community with challenges referenced on the site; not itemised here.
- Microsoft’s AdventureWorks sample is used throughout Learn and the star-schema article.
6. Communities
- Data Visualization Society (volunteer-run 501(c)(3) non-profit; Slack community; Nightingale publication; DVS challenges; confirmed 2026-10-08 via WebFetch, curl gets 403).
- Microsoft Fabric / Power BI community (https://community.fabric.microsoft.com/powerbi) and Power Platform community (https://powerusers.microsoft.com/), linked from Microsoft Learn.
- MakeoverMonday and Workout Wednesday threads on social media.
- Metabase community forum (linked from Metabase Learn).
7. YouTube creators
Channel pages could not be opened (consent redirect), so descriptions come from search results.
- Guy in a Cube: Adam Saxton and Patrick LeBlanc; Power BI and Fabric content (per search results; Microsoft employment not verified).
- Curbal: Ruth Pozuelo Martinez (Curbal AB, Microsoft Data Platform MVP, per curbal.com 2026-10-08); 800+ videos, free intro Power BI course, blog at curbal.com.
- Alex the Analyst: SQL, Excel, Tableau, Power BI and Python playlists (per search results/Class Central listings).
- SQLBI videos (https://www.sqlbi.com/tv/): DAX and modelling depth.
- Tableau, Qlik and Looker official channels: not checked.
8. Learning BI with AI assistants
Use ChatGPT, Claude, Gemini or Copilot as a tutor and pair-programmer, not as the source of truth.
- Useful (author’s advice): explain a DAX/SQL/LookML snippet line by line; translate between SQL and DAX; generate practice questions on a dataset; draft routine measures (simple aggregations, basic CALCULATE filters, standard time intelligence); refactor and comment code. SQLBI’s article “Where we stand on AI and DAX in 2026” (2026-09-22) states AI is now good at writing common DAX.
- Caution 1, plausible but wrong: SQLBI says most wrong DAX is syntactically valid but returns a subtly incorrect number through filter context, context transition or relationship interaction, and the failure is invisible unless you can reason about the engine. Validate against a known total at several granularities.
- Caution 2, model problems: the same article notes AI will write a long measure to work around a missing relationship or badly shaped fact table. Fix the star schema first.
- Caution 3, learn the concepts: SQLBI argues training now matters for filter/row context, context transition and validation rather than syntax. (The article embeds commentary attributed to AI models; treat its figures as opinion and read it yourself.)
- Microsoft’s own docs: Copilot in DAX query view (https://learn.microsoft.com/en-us/dax/dax-copilot, updated 2026-06-29) can write, explain and run DAX queries, and says newly added DAX functions may be unknown to it, so explanations may be incorrect; it works only from model metadata, needs a Fabric capacity workspace, and checks syntax with one automatic retry.
- Copilot for Power BI (https://learn.microsoft.com/en-us/power-bi/create-reports/copilot-introduction, updated 2026-08-31): as of 2026-10-05 the report Copilot pane is generally available while the standalone Power BI agent and Copilot in apps are preview. It requires paid Fabric capacity (F2 or higher) or Power BI Premium (P1 or higher); trial capacities and free SKUs are not supported, and Pro or PPU alone is not enough. So a free-tier learner usually cannot use it; use a general chatbot instead and paste the schema (never confidential data).
- For LookML and SQL: the same discipline applies (paste a small table/field list, ask for the query, run it, compare with a manual count). No Looker-specific AI learning guidance was verified in this job (the Gemini-in-Looker docs URL tried returned 404); see looker-learning-path.
- Practical rule: ask the assistant to explain its answer and propose a test, keep a personal “known-answer” dataset, and never paste company data into a consumer chatbot without approval.
9. Vault video summaries
None found. The yt-base raw layer (/mnt/d/OBS/yt-base/) was not mounted in this environment (WSL has no /mnt/d), so no creator summaries could be searched. Searching the vault (KB-*, RESOURCES) for Guy in a Cube, Curbal, SQLBI, Alex the Analyst, Enterprise DNA, Kimball, star schema, Makeover Monday and Storytelling with Data returned only KB-TECH/DATABASES/data-mapping-template.md (not a video summary). Re-run when /mnt/d is available.
Sources (fetched 2026-10-05)
- https://learn.microsoft.com/en-us/credentials/certifications/data-analyst-associate/
- https://learn.microsoft.com/en-us/training/powerplatform/power-bi/
- https://learn.microsoft.com/en-us/power-bi/guidance/star-schema
- https://learn.microsoft.com/en-us/power-bi/create-reports/copilot-introduction
- https://learn.microsoft.com/en-us/dax/dax-copilot
- https://www.sqlbi.com/ and https://www.sqlbi.com/books/
- https://www.sqlbi.com/blog/marco/2026/09/22/where-we-stand-on-ai-and-dax-in-2026/
- https://www.kimballgroup.com/data-warehouse-business-intelligence-resources/books/data-warehouse-dw-toolkit/
- https://www.storytellingwithdata.com/books
- https://clauswilke.com/dataviz/
- https://www.makeovermonday.co.uk/ , https://www.workout-wednesday.com/
- https://www.datavisualizationsociety.org/ , https://www.kaggle.com/learn
- https://learning.qlik.com/ , https://training.thoughtspot.com/ , https://docs.thoughtspot.com/cloud/latest/
- https://www.skills.google/paths/28 (WebFetch), https://cloud.google.com/blog/products/data-analytics/looker-certification-program , https://cloud.google.com/learn/certification , https://docs.cloud.google.com/looker/docs/build-skills-with-courses
- https://trailhead.salesforce.com/content/learn/modules/cert-prep-tableau-data-analyst , https://learning.qlik.com/student/activity/2161345-qlik-sense-data-architect-certification-exam-details , https://www.sqlbi.com/p/introducing-dax-video-course/ , https://curbal.com/ , https://www.datavisualizationsociety.org/ (second pass 2026-10-08)
- https://www.sigmacomputing.com/go/hol , https://help.sigmacomputing.com/
- https://www.metabase.com/learn/ , https://superset.apache.org/user-docs/intro/
- Web searches (secondary): Tableau certification and training; Looker courses; Tufte/Few/Cairo/Wilke book details; Guy in a Cube, Curbal, Alex the Analyst channel descriptions.
Re-check 2026-10-07 (verification pass)
URLs that loaded (HTTP 200) with content checked: Microsoft Learn PL-300 certification page (100-minute proctored exam, free renewal by online assessment, practice assessment and sandbox; metadata updated 2026-04-21), Power BI training hub (updated 2026-06-26), star-schema article (updated 2026-04-01), Copilot for Power BI intro (updated 2026-08-31; report agent Copilot pane GA, standalone Power BI agent in preview, paid Fabric capacity F2+ or Premium P1+, trial and free SKUs unsupported), DAX Copilot (updated 2026-06-29; Fabric capacity, model metadata only, one automatic syntax retry), SQLBI 2026-09-22 article (evaluation-context wording confirmed), SQLBI books page (Definitive Guide to DAX, Third Edition), Kimball Group Toolkit 3rd Edition page, Claus Wilke’s online book, Makeover Monday, Workout Wednesday (tracks Tableau, Power BI, Sigma, CRM Analytics visible; active weekly through 2026 Week 38), Sigma hands-on lab (five guided modules, free trial), Qlik Learning (free just-in-time videos and select courses, subscription tier), Looker certification announcement (Looker Business Analyst and LookML Developer), Kaggle Learn, Metabase Learn, Superset user docs, dax.guide, daxpatterns.com, powerusers.microsoft.com.
Returned 403 to the fetcher on 2026-10-07 (existence not re-confirmed): tableau.com/learn, skills.google/paths/28, datavisualizationsociety.org, community.fabric.microsoft.com/powerbi, curbal.com.
Not checked in this pass: Tufte/Few/Cairo/Knaflic book details, Alex the Analyst and other channel descriptions, SQLBI “550 videos and articles” and “Introducing DAX free” claims (not matched in fetched text), Looker current exam names, ThoughtSpot University course titles, Sigma Academy.
Open items
- 2026-10-07 pass: items under “Not checked” and the 403 URLs remain unverified; agy-search timed out and WebSearch quota was exhausted.
- tableau.com/learn is blocked (403 on all routes); Tableau free training page contents not verified. Superset docs root returned 403/empty on 2026-10-07 (not retried).
- Looker certifications are from a 2020 announcement and not on Google’s current certification page; fee unknown. Tableau and Qlik exam facts now come from Salesforce Trailhead and Qlik Learning pages (second pass).
- Tufte, Few and Cairo book details came from search results, not publisher pages.
- Kaggle SQL course, Sigma Academy, Enterprise DNA, Tableau/Qlik/Looker YouTube channels not verified.
- YouTube channel pages not opened (consent redirect).
- Gemini-in-Looker and other vendor AI-for-learning docs not verified.
- yt-base not accessible; vault video summaries not searched.
- power-bi-learning-path and tableau-learning-path were written by parallel jobs and linked here after they were confirmed to exist (2026-10-05).