Methodology to identify Data Products: Our approach is, to focus on Data which is relevant for value creation. How can we identify those candidates with ONE DATA and/or our Data Catalogue and/or with Cartography?
Hm, I am not fully sure if I got your question right, but I will try And this is just my point of view on the current situation, I do not have a deep technical grasp yet.
For the Data Catalogue DC, the data that is focused is funnily the data that is already there in a given ONE DATA instance. However, with an increasing size of an instance, more and more data is technically available, so someone starting up a new project might simply just not be aware that a for him/her useful dataset is there already for the taking.
And what is the consequence of them not finding that data? They create a new dataset instead, leading to many redundant datasets all over the place. Ignoring the fact, that it takes a lot of time to create those (usable) datasets in the first place.
So I’d say, the data the DC focuses on is already existing data, while the value generated out of the data product is transparency and awareness in the first place. Furthermore, the DC produces several important information like samples of the datasets I can look into, statistics, and especially the data lineage (so where does the data come from, what changed it, etc.). All of this information makes datasets better usable, as users can more quickly get an understanding about them.
The Data Cartography DCart on the other hand is something that transitively pays into the DC and its value by filling it.
The DCart is also based on data already existing in an OD instance and it gives automated possibilities to fill a Data Catalogue. What I’ve mentioned before with the growing OD instance and its datasets can also quickly be overwhelming - a DC needs constant care and maintenance in order to fully be effective.
So the value generated by the DCart is help filling a DC, whose value I’ve described above.
Hope my thoughts are useful for you, I am very interested in what you think
I’ve left out OD now, as it will need a waaay longer answer
P.S.: I also took the freedom to move your question to the “data products” category, hope that is fine for you!