Source the data in a dataviz solution

 

Source the data in its dataviz solution

What is hidden behind a variable, an expression, a simple dashboard cell? This is not necessarily easy to know, especially if the dashboard is complex.

Methodology:

We offer a fine and automated introspection of certain dataviz solutions by an automated scan of dashboards: templates, and metadata, in SaaS mode.      analysed technos:

Granular analysis   openAudit analyzes the structure of the dashboards, the metadata found there, therefore the intelligence, the structure, the sources, etc.   Low impact    This parsing is done at the "delta" so as not to overload the servers. All the metadata collected is centralized and logged.    Continuous analysis   openAudit only continuously analyzes dashboards that have evolved. A unit versioning takes place daily on a git.

Output :       Data lineage down to feed layers

1- A search engine to isolate a datapoint from the dashboard. You enter the name of the object, block, filter, variable, the expressions… etc, in a simple search engine which proposes by autocompletion all the related answers.   2- An instant display of data lineage   All the transformations are displayed instantly up to the physical sources. The data lineage can possibly be extended in the databases (Cloud, on prem') to the operational sources!

 

NB :  We also offer impact analysis grids which define, from a field, the impacts in the dashboard, Replicated dashboards can be identified, All the data that is produced but not used in the data visualization layer is identified (directly or via expressions, variables, filters, etc.), In PowerBI, the process includes the data lineage of the SSAS parts (such as Power Query M, SQL queries, calculated expressions, DAX), but also the MDX language and the XMLA protocol.