Four issues Data Warehouse could solve!

Abhilash Marichi
2 min readSep 11, 2020

--

Photo by Lukas Blazek on Unsplash

I am assuming that you already have a working system that stores data in some way and if you are facing the below issues then a Data warehouse could solve it.

Issue #1 — Your system does not store historical data.

If your system always updates the data and does not have the capability to maintain the history then Data Warehouse could help you to store the point in time data.

For example — Let’s say your system captures the Profile of the user which has their place as one of the attributes and John is one of the users who was living in Chicago as of Jan 2018 and moved to Minneapolis as of Jan 2019 and has again moved to Kansas City as of 2020. If this is the case, your system would have only the current place of John. If you build a data warehouse and load the data into Datawarehouse regularly it can keep the point in time information and you can use it for your business needs.

Issue #2 — Your system has multiple versions of Truth

For this, let’s assume your business runs two separate systems that stores user profile information similar to the above example, and let’s say John is regularly updating his place in one system whereas he is not updating his info in the other system. Now when you generate business reports out of these two systems one system says John is in Kansas City and another system says John is in Chicago and you would get confused which is True. A data warehouse could resolve this by maintaining the integrated data in one place and providing you the right information.

Issue #3 — Your systems are scattered across different technologies or platforms

If you are running multiple systems, similar to the previous example, and if you running your systems on different tech stack for storing data then for your business needs a Data warehouse could be built by integrating all of the systems.

Issue #4 — Your business reporting has poor performance

If generating your business reports is a pain and it is taking days or weeks to generate a business report then you should definitely consider building a Data Warehouse that is because it is built on for OLAP (Online Analytical Processing) which means a Data warehouse always makes retrieving of information damn easy and it improves the performance when you generate business reports.

A data warehouse could be used for many other purposes also, like Data Mining, various forms of Analytics, and Machine Learning.

Hope this gave you a brief idea of what a Data warehouse could do and let me know if you are facing any of these issues!

--

--

Abhilash Marichi

Data Engineer at Amazon. I write about Data, Product & Life.