# Re-Upload - Tableau Next Semantic Modelling [Reposted]

> This is content from just-tim, the data-and-analytics channel by Tim Ngwena (formerly 'Tableau Tim'). Tim has 12+ years of hands-on BI experience and covers Tableau most of all, plus Power BI, Looker, Hex, SQL and data modelling, the analytics industry, and the craft of doing the job — always tool-agnostic and honest about the trade-offs.

- **Author:** Tim Ngwena (just-tim, https://just-tim.com/about)
- **Published:** 2026-04-10
- **Format:** Video · 51 min watch · transcript available
- **Topics:** Data prep, Analytics
- **Tools:** Python; Tableau (data modelling, relationships, semantic models)
- **Canonical:** https://just-tim.com/posts/re-upload---tableau-next-semantic-modelling-reposted
- **Watch:** https://www.youtube.com/watch?v=PQr2-oFaIvs

This is day two on my new channel, where I walk through building a semantic model in Tableau Next using the classic Bookshop dataset. I cover uploading data, the difference between data lake objects and data model objects, building relationships and unions, and testing the finished model — bugs, detours and all.

## Key takeaways

- Semantic modelling in Tableau Next is powered by Data Cloud and can be done either in Tableau Semantics or directly in Data Cloud — they're part of one experience.
- Data Lake Objects (DLOs) are raw, source-aligned storage while Data Model Objects (DMOs) are a business-friendly harmonised layer on top, roughly analogous to underlying CSVs versus a single Tableau data connection.
- Unions are built inside a logical view, but you must manually map matching fields one by one — a clunky experience compared with Tableau Prep or Desktop.
- Calculated fields can be created on a data object (e.g. STR(bookID1)+STR(bookID2)) to fix join keys, but they don't carry through into logical views as I'd hoped.
- You can test a model by picking dimensions across tables; the resulting semantic query is JSON metadata that can be converted to SQL against your source.

## Chapters

- 0:17 Plan: where semantic modelling happens
- 2:05 Uploading Excel and a vibe-coded CSV splitter
- 4:43 DLO versus DMO terminology
- 6:18 Creating data lake objects through the bugs
- 14:08 Starting the semantic model and first relationship
- 19:23 The info table calculated-field blocker
- 24:14 Building the remaining relationships
- 30:24 Building a sales union in a logical view
- 32:38 Wrestling with matching fields
- 41:22 Connecting sales via ISBN
- 44:10 Calculated fields for the book ID join
- 47:38 Testing the semantic model

Watch the full video, read the transcript and use chapter deep-links on the page: https://just-tim.com/posts/re-upload---tableau-next-semantic-modelling-reposted

---
just-tim — Data and analytics, with a point of view. · https://www.youtube.com/channel/UC7HYxRWmaNlJux-X7rNLZyw · https://twitter.com/TableauTim · https://www.linkedin.com/in/timngwena