Digital transformation is not as clean as diagrams make it out to look. Not that the diagrams are clean in the first place. I did find a pretty clean one that I love in "Beyond Big Data Basics: In-Stream Processing Cures Batch Processing Blues" on iamwire.com shown below. This diagram is clean because it is not a detailed infrastructure diagram, it is an explanation of how stream and batch data can be combined with data science for positive improvements across a business. One of the things I love most about this is that it inadvertently articulates some of the non-marketing needs for customer data platforms.
There is room in the original diagram for a customer data platform to help organize this information; to allow channels to interact with, collect, and make use of this data. Imagine inside-out parentheses overlaid on the diagram to the sides of the data lake, so that the arrows hitting "online applications" and "technical support" are all intersected by the parentheses )( ... or don't. Let me show you what I mean: