There is a category of person who needs data training and is served terribly by the market: the operations manager, the marketer, the founder's first business hire, anyone whose job involves answering questions with numbers but who is not going to become an engineer. Most free material aimed at them is either a fifteen-minute tool walkthrough that teaches button positions, or a data science course that opens with Python environments and loses them in week one. Metabase Learn is aimed squarely at that person and does a better job than the tool's ownership of it would suggest. The vendor-neutral half is the good half.
The Learn SQL track teaches query writing and, more usefully, query debugging, which is the skill nobody teaches because it is harder to write a tutorial about. Database Fundamentals covers the concepts a non-engineer needs to stop being confused: what a join actually does to row counts, why a table has a primary key, why the number changed when someone added a filter. The Data Landscape track explains warehouses, ETL and where data comes from inside a company, which is context that analysts usually acquire by accident over two years. Business Analysis Methods covers funnel analysis and trend discovery with actual reasoning about what makes a funnel comparison valid.
Spreadsheets to Databases is aimed at the very common situation of someone who is excellent in Excel and stuck there. The Metabase-specific half is a competent product manual and not much more. Getting started, building questions and dashboards, administering an instance. Fine if you run Metabase, skippable otherwise.
The cheat sheets sit between the two halves and are better than they need to be, particularly the chart selection guide, which is the kind of one-page reference that stops someone putting eleven categories in a pie chart. The gaps are worth naming. There is no certification of any kind, which for this audience is a real loss. The analyst who works through a data literacy curriculum has nothing to point at afterwards, and this is precisely the population that benefits most from a credential because they are usually arguing for a role change rather than applying with a relevant degree.
A free badge would cost Metabase very little and would be worth a lot to the learner. Depth stops early by design. Anyone who already writes SQL competently will find nothing new, and there is no intermediate track to move up to, so the material is a launchpad rather than a curriculum. The more interesting omission is governance.
Self-service BI has a well-documented failure pattern: adoption succeeds, everyone builds their own questions, and eighteen months later there are six definitions of active user, four dashboards claiming different revenue figures, and no way to tell which is right. A tool that makes querying easy makes that problem arrive faster. The material teaches people to ask questions and says almost nothing about the discipline of agreeing what the answers mean, which is the actual hard part of running analytics in a company. Metabase is not uniquely guilty here, every BI vendor has the same blind spot for the same reason, and it is still the thing I would add if I were writing the curriculum.
Worth knowing as context: Metabase is open source in its core edition with paid Pro and Enterprise tiers, and the split between what is free and what is not shifts over time in the way it usually does with open core products. The learning material does not make that boundary clear, so an analyst following a tutorial can reach a feature that needs a licence without being told in advance. Minor, and annoying. Four point three.
A free, well-judged data literacy curriculum hiding inside a product manual, aimed at the right audience and let down only by the absence of any credential to show for it.