when I thought my teacher was the best , then I found James Oliver....wow ....what an explanation ......wow ......i understood it from a better perspective now ....Thank you Sir James
I'm returning with more knowledge than previously...lol Wouldn't it be beneficial to start with the business process? How were you able to run the new environment in parallel with the existing, if you are moving the data? I think I'm getting something mixed up here.
Hi James, Nicely done, how you explain step-by-step what is going on in a project. This shows you need to understand business requirements, the architecture which needs to be migrated and the data warehouse architecture, where you want to integrate.
Hi James, thank you so much for your thorough and detailed explanations and analogies. I really found your video helpful. One question I have on my mind, how long did it take for you to complete building and designing the DWH? What was the overall time frame for this project to deliver this solution to business?
Great question. This was done under a compressed timeline that not only included building the data warehouse but a new reporting environment and making the users comfortable with the new environment through training. The total time was approximately 6 months. However, under normal conditions, this is extremely aggressive but I had no choice so there were many sleepless nights and long weekends. Because I was the main architect doing most of the technical work I was able to save time by cutting out overhead and providing all the direction to the DBA and ETL teams.
What was total project time from spec to users using the EDW ? Intrigued because I'm about to start building a dwh for the org I work for. That tip about the views was genius! much easier for devs to have something to shoot for that's in front of them, compared to doing discovery work.
Sorry for not answering sooner, missed this one but I would say make sure that the subject matter experts (ie those that understand the data best) are made to feel like they are part of the project. When building a data warehouse inevitably the devs will have many questions about the data and they cannot be waiting on people for answers or it can significantly impact the timeline. Do the developers have prior data warehouse experience?
@@JamesOliver The org is quite fresh to modern data practices. Undeniably the devs will be enrolled in all the conversations. Its green fields here, likely we would bring in extra skill required. My big concern is the business doesn't have a clear picture on what end state should look like. I have my own vision for it, just wanted a baseline for total time if alignment is succinct across the biz
Yes it can be a challenge for the business to wrap their heads around what this is going to do for them. Funny enough I’m starting up a new Edw project myself and I presented to leadership and made sure I used examples they could relate to. For example, are there a specific report request for an executive that took forever to deliver because IT had to go hunting down the data? Let them know this will bring these requests down from weeks to days or months to weeks with a data warehouse. That should help.
