Thinking Beyond the Model

Why is it that people think that using the Data Vault model is good enough? You won’t gain much benefit from the model without the methodology anyway. That was the secret sauce behind even successes in 1.0.

However, 1.0 has become old and can’t deal with many issues in the current landscape of applications, software and data.

That is why Data Vault 1.0 was just a type of Data Warehouse architecture and was competing with other architectures in that space. Data Vault 2.0 encompasses a completely new way of thinking where it’s actually a complete system of business intelligence and deals with everything including people, process and technology.

There’s a good reason why people are having issues with Data Vault 1.0 projects these days, but DV 2.0 has a plethora of success stories and have saved customers a ton of money in projects. They’ve been presenting these successes at the annual world wide data vault consortium for the last 5 years which is the only Data Vault focused event in the world.

It’s important to understand that DV 1.0 is very focused on the data warehouse model only and has limited applications in today’s environments. You can probably still make 1.0 projects a success if you follow the methodology correctly, but it would be too challenging in today’s environments. It’s very easy to make mistakes and veer off the path and break the architecture. You’d also lose out from the new innovations and best practices. That is why whenever we hear about difficulty or failures on Data Vault projects, on further investigations it has usually turns out to be a DV 1.0 implementation and even more when someone didn’t follow the rules. For some weird reason, customers are willing to dump money trying to fix their broken DV 1.0 implementations, rather than to actually take a look at DV 2.0 which would have helped. Of course, even DV 2.0 with only the model focus is a sub-optimal way of looking at it. I really don’t recommend working with such customers. Not only are they shooting themselves in the foot, they’ll take you down with them. (Note: In some instances, you as a consultant can end up serving as a scapegoat as well, so be aware. More on that another time).

Data Vault 2.0 is a solution based approach where everything is already prescribed for you, and has been proven with pretty large scale solutions across various verticals. Naturally, the savings tend to rack up when you follow such proven best practices. To give you an example, one consulting firm presented a success store in DV 2.0 as far back as 2013 where they saved their customers around $25M per year, which is no small feat. And, this was on a NoSQL platform. Unfortunately, I can’t give more details on this at the moment, but if you attended WWDVC 2013, you’ve seen this case study.

And, it’s been around for 5+ years now. It’s a good time to say goodbye to Data Vault 1.0.

Think about it. Would you still use Windows 95 (1995), 98 (1998) or NT (1993) or Mac OS 9 (1999), 10.2 (2002) today? Data Vault 1.0 was developed in 1990 and released in 2001 to give you an idea. Dan even stopped certifying people in 1.0 around 2008. Data Vault 2.0 was first announced in 2013, but certification didn’t start till much later. It’s the latest in Data Vault. It gives you a full spectrum solution where the model is just a component which is the way it should be and there’s a large base of people who are already reaping the benefits.

Anyway, the model is just the starting point even for Data Vault 2.0 and it will give you some benefits with the way it integrates by hub and decouples by satellites etc. Stopping at the model is not the right way to approach it because the true benefits lie in when you go beyond the model.

You can learn more about in this article that Dan wrote:


If you start comparing DV 1.0 with DV 2.0 and only compare the model portions, you are doing yourself a big disservice. Both of these are invented by the same author which would be reason enough to look at 2.0 which is a major evolution.

Data Vault 1.0 served us well … till about 2012-13. After then, it has always been the right time to evolve and look at Data Vault 2.0