Data Analytics: Does one size really fit all?

A plethora of terminologies, acronyms and technologies are currently trending in the world of data analytics. For both the un-inaugurated and the seasoned professionals it has become a minefield, open to misinterpretation and misrepresentation.

Are you confused about EDW, DW, Data Lakes, Hadoop, HTAP, Hive, LLAP, Spark SQL (you can complete the list). Are you wondering ... To ETL or not, to ELT or not, or to do no ETL/ELT at all?

Authorities on the subject of Data Analytics are beginning to agree: there is no one size fits all.

Safe-routes are chosen

Amidst several new entrants to the market and during the incubation (transition period to modern data warehousing), everyone in the know (including those who think they know!) is taking the approach of combining only certain portions of both old and new methods. Proposing all of Business’ data problems will be solved. What can go wrong? It is a safe route and makes sense while testing the waters.

Myths, Fables and Falsehoods

Unbeknown to the Business User, acquiring more technologies simply raise the levels of the already complex Analytical / Business Intelligence / Data Warehouse / Data Integration landscape. We are enticed into falsehoods of: ease of use, data will be prepared in days rather than weeks and self-service solutions. Hurray, no need for an IT department!

Myths such as “Detailed data is not available”, “It takes too long to load data”, “Too much effort to incorporate new data” and so forth is being advocated as reasons to divorce anything traditional and to dive headlong into new lakes in the cloud. Ask pertinently how data is categorized, defined, cleansed, time-lined only then to be advised of having to incorporate existing strategies to address these traditional and still valid challenges.