If you want to have a good quality data in your DApp, it will be all about setting up metadata standards. Sure, it's easy to pass data around as arbitrary JSON blobs, but if you later need to build analytics on top of this data or share it with a 3rd party, you need to enforce data consistency and interpretability.
In relational databases, this is usually done with DB schema modelling, where you set up constraints on data. This helps with data consistency, but does little for data interpretability. This is where the semantic graph database, such as Fluree, comes to help. Not only you can set same type of constraints you'd set in a relational database, but you can also adopt global semantic schemas, that power search engines of such tech giants as Google and Microsoft. You will be able to define ontologies on top of your data, explaining how one entity is related to another entity. This can make your data really tell fascinating stories not only to you, but also to your partners, thus giving you a competitive edge.
Listen to Fluree introduction by Brian Platz, where he dedicates significant amount of time to various metadata standards and methodologies.
Also take a look at this post about the Semantic Interoperability in Fluree Data-Centric Blog series