DATA KNOWLEGE
Starting point of managing Data is to know what Data you are speaking about and according to , what are the Data you manage . Amongst the Data a company is dealing with there are the ones (Standardised/ recurrent) which are designed for delivering the products and services (from the business processes) and all the others , not planned, not structured (not standardised) like mails, sheets, slides, photos, audios done for supporting business activities management. Both are in the scope of Data governance and need to comply with Data governance rules.
Standardised business data are the ones which are key when speaking about company business model, effectivity, efficiency , digital transformation, Data assets and this is on them that company Data knowledge needs to focus on.
Knowing what are the standardised business data you manage (yours and the third party ones) relies on 2 parts:
1/ The business Data knowledge;
2/ The business Dataset knowledge.
The business Dataset knowledge(2) is the one any company already knows and which is managed by your IM function, identifying which dataset is in which application/platform with an IM semantic. It should contain the dataset cartography, but to elaborate a cartography you need to know also to what these datasets are linked to make it useful and actionable.
This is where the Business Data knowledge(1) is becoming crucial and often not done as it is under all the businesses responsibility and competences to do it (this is one added value of Data governance to federate and support all these businesses in doing it) . It relies on 3 items:
Business Object dictionary (definition/ontology) , Business Object model (BO links) and the Business Object Catalog.
All using a business semantic allowing any employee to easily understand the content, purpose etc facilitating their use for other purpose and then develop new business opportunities. The business catalog is describing for a given BO the different business contexts the BO has been used allowing to establish the cartography with the Datasets (including quality and source of thru).
Figure below illustrates Data knowledge content
Without this global Data knowledge (which will be progressively recovered and never exhaustive) no effective digital transformation.
In addition the BO dictionary should not be too company specific (we never work alone). It is more “industry specific” and we should be able to find standardised industry BO definition/ontology allowing much more efficient data exchange / collaboration when using common European Dataspace for example.
This is part of next decade challenge