The named ideas behind the model — 14 metadata concepts, grouped by family. Each one is a small, self-contained idea you can point at.
Show the structure, don't just describe it — generated diagrams, whiteboards and visuals that let everyone see the same model and correct it.
Metadata lives in version-controlled files — reviewed, tested and deployed like code, so the model is an engineering artifact rather than a GUI state.
Left-to-right flow vs top-to-bottom drill-down.
Lineage traversed backwards: what breaks if I change this? The question every engineer asks before touching anything, and the one most platforms cannot answer.
Data lineage made clear and actionable for decision makers.
Clear, consistent, metadata-driven diagrams that connect business and technology.
Related business concepts mapped together as a connected constellation, not a list of isolated entities.
A generated SVG carries links, so the picture stops being a dead end and becomes the navigation layer of the model.
Generate the building blocks, not the picture: a shape library from the model — so a workshop can stay interactive while every block on the canvas is a real concept.
Model with the business, don't model for them — sketch on a shared canvas until the picture matches how they actually think. The diagram is where agreement happens.
A diagram generated from the structure is always current; a hand-maintained one rots. Read the picture from the source, don't transcribe it.
The anti-pattern: understanding locked behind a person, a licence or a login — access controlled by whoever holds the key rather than by whether you need it.
The anti-pattern: treating the model as a picture to be drawn rather than knowledge to be captured — so it informs nobody and generates nothing.
The other half of structure — making the model seen, not just real. Structure makes knowledge exist; visual thinking makes it visible enough to trust, question and act on.