Finding your Data Architecture

If you’re a data architect – you’d recognize the different approaches (and names) to data warehousing such as Ralph Kimballs star-schema model, or Bill Inmon’s normalized approach, or even Dan Lindstedt’s data vault. But if you’re like most people, you’ve never even heard of these names and approaches. Which data architecture approach should you select as a business owner?

Like the answer to any good tech question – “it depends”.

Most small businesses can probably get away with using any of these successfully. I’d say a good rule of thumb is to store information in a data warehouse and build star-schemas to be consumed in some type of visualization tool (PowerBI being my tool of choice). FWIW that would be a mix of Inmon/Kimball approaches with a “must-have” on the Kimball star-schemas.

Here are some questions worth asking when considering your data architecture:

How many sources of data will be combined into this single area?

How is this information going to be consumed?

Who is the audience?

Do we have resources in house that can maintain this and build on top of it?

How often do we need the data to be refreshed?

What tools does this data need to work with?

Building standards around your data architecture needs to be a full picture and needs to fit within your existing toolsets as well as new ones that can/will be added. It’s not an easy task to build out your data architecture but if done correctly will set you apart from your competitors.

A Daily Email From Me

Give me a couple minutes of your day and I promise to provide thought-provoking messages to help you level up your BI knowledge and run more successful data projects.

A Daily Email From Me

Subscribe to get daily tips/ideas for how to best run a data project.

    We won't send you spam. Unsubscribe at any time.