If you did have the opportunity to work on a BI project, let him give you an overview of the phases of this kind of project:
- Phase 1 – Ingestion of the data into a Data mart or big data environment – During this phase, we bring the data from multiple systems and load it into a Data warehouse, cloud, or big data
- Phase 2 – Transformation – During this phase occur the transformation or calculation of the metrics, here can apply techniques like snowflake schema, normalization, and de-normalization of the data
- Phase 3 – Reporting or Analytics – During this phase, the integration between multiple data sets will happen to populate a report or populate results in an analytics app
During these 3 phases, we can do different types of automation, typically during phase 1 we have the easiest approach for automation, for phase 1 typically we just need to compare the source against the target, this type of automation will be called monkey automation, due to you not need to think a lot, you just need to make sure that the data match.
For Phases 2 and 3, he wants to introduce the concept of Automation flow, meaning one of more automation suites merged together to test the application in an E2E (End to End), but how we can determine that Automation flows? The answer is industrialization, if you know how your application works on the business side, this will going to give you the visibility to create those automation flows.
Those automation flows can contain multiple subject areas needed to certify that the transformation is accurate and identify the potential opportunities areas during the transformation process, before even testing the reporting part.
Takeaways from the topic:
Understand the importance of industrialization in the career path of a test engineer in the BI world, and how having a good understanding of the industry help to create an intelligent automation suite. How to increase the scope of the automation using functional flows.
Jose Adolfo Lozano Marín
