Semantic modelling
You can create a semantic model utilising the data objects you have saved in your workspace.
Doing so will allow you to create relationships between your different data objects.
Creating relationships follows a similar interface in Tableau Next as it does in Tableau Desktop. You will still need to set what the matching fields are in both data objects.
Always click apply after each relationship field is set!
Tableau Next also has a very useful hack of allowing you to bring in chosen columns from any of your related data objects, so you can test to see if the data model is working fine. Then from this you can create a semantic view and create visualisations.
Why use a semantic model? The pros:
It acts as a user-friendly interface between messy databases and a business. It can simplify table names into easy to understand ones for clients, you can create relationships - so it is easy to understand how the tables are connected, and you can set metrics, so everyone uses the same calculations for that metric.
Semantic Models are very beneficial to AI Agentforce; when it is asked a question about how the company is performing, it has the semantic model to use to come to a precise conclusion - of course, only if the semantic model itself is clear and usable.
The Cons:
If you are working with a large database with many large and different tables, creating a semantic model for all of them could be quite hellish. Figuring out how each data object (table) relates and standardising different metrics would definitely take a lot of time.
But once that is achieved, the benefits are endless!
There is also the cost issue. The pricing is quite steep and Tableau Next is only available to customers that have premium Tableau+ licenses.
The Take Home?
Tableau Next is pretty useful for big businesses and semantic models can help new analysts understand a businesses data model easier!