Avoiding the Pitfalls of Data Warehousing: A Spotlight Q&A with Gertjan Vlug and Steve Aspris of BIReady

Originally published 16 January 2012

BeyeNETWORK Spotlights focus on news, events and products in the business intelligence ecosystem that are poised to have a significant impact on the industry as a whole; on the enterprises that rely on business intelligence, analytics, performance management, data warehousing and/or data governance products to understand and act on the vital information that can be gleaned from their data; or on the providers of these mission-critical products.

Presented as Q&A-style articles, these interviews conducted by the BeyeNETWORK present the behind-the-scene view that you won’t read in press releases.

This BeyeNETWORK spotlight features Ron Powell's interview with Gertjan Vlug, CEO and President of BIReady, and Steve Aspris, Managing Director of BIReady LTD in the UK. Ron, Gertjan and Steve discuss the advantages of data warehousing, the BIReady model-driven approach, how BIReady interfaces with CA ERwin, and their partnership with Teradata.

Steve, one of the commonly stated concerns when an organization considers a data warehousing implementation is that a data warehouse is very expensive and time-consuming to build. As a result, many organizations try to achieve data warehousing functionality without a data warehouse. Can you explain the pitfalls in this type of approach?

Steve Aspris: Yes, nowadays there are some very good front-end business intelligence (BI) tools that will allow you to report on data straight from your live operational systems. There is also a current trend toward what is called data virtualization, which again builds views on live operational systems, providing a snapshot of the current business. In contrast, a data warehouse handles history – bridging across the critical dimension of time. For example, it can report on sales by year, then drill through to sales by quarter and still preserve the correct version of the customer's address both at the time of the transaction and as of the current date. In that sense, the data warehouse very elegantly handles and reports on business change as it occurs over time. We all quickly experience how business is never static. When you think about it, after death and taxes, business change is probably the third thing that is guaranteed!

The second major advantage to having a data warehouse in place is maintaining a single version of the truth, what Bill Inmon and David Loshin call the System of Record. Combining all business intelligence into a single repository solves many issues on data quality. For example, having a single authoritative customer list means different reports will not contradict each other because they were drawn from the same integrated and conformed data source; and that data source, of course, is the data warehouse.

These are just two examples; and, of course, there are many critical advantages to having a data warehouse in place. The pain felt by organizations that try to build one is what gives data warehouses a bad rap, and reasonably so. Because stacked against these advantages are huge cost and complexity issues. Building a data warehouse by hand is expensive – very expensive, very slow and very complex. Even after the enormous investment of manually building a data warehouse, if the project even survives long enough to reach the maintenance and growth phases, companies quickly find out how brittle and inflexible the system is to change. This is what BIReady solves – taking out the cost and complexity by using a data warehouse management system to build it for you.

Well let's get into specifics. What is the value proposition of BIReady? You stated earlier that it takes out complexity. Can you elaborate on that?

Steve Aspris: Well, about a century ago, Henry Ford proved to us that automation is the key if you want to drive down costs, speed up production, and produce consistent repeatable results of the same quality. It's really no different in information technology. For decades now, information technology has been taking slow, expensive, error-prone tasks and building software that automates those processes. This is the value proposition.

Business intelligence has matured, and we know what the repeatable best practices are. BIReady builds the data warehouse by enforcing these best practices in a consistent and repeatable way: connecting to your source systems, building a relational warehouse, and implementing star schemas for reporting. Critically, BIReady does this in a few minutes instead of weeks of manual work for a team of developers. The value, of course, is highlighted in today's business environment. In this financial climate, we are in a perfect storm. Business leaders need better reports. They are demanding high quality business intelligence. And yet, at the same time, budgets are tighter than they have ever been. So the expensive route of building a solution by hand is just no longer the option.

What typically, Steve, is the time to build a data warehouse with BIReady? Are we talking years? Are we talking months? Are we talking weeks?

Steve Aspris: When we think about a data warehousing project, it's a business problem that requires both a business and technology solution. Consequently, the business side of it will take the same amount of time. You still need to analyze how the business works, and you still need to build a conceptual model. It's really the implementation side of things where BIReady consumes that model and kicks in to bring value. We often demonstrate, for example, where we're able to take some data through the entire lifecycle from its source system to a useful destination data mart and create an OLAP report within 30 minutes. That’s an example of how quickly we can handle the technology aspect of the solution. Of course, what we're doing in real life is taking the entire process down by about 50% in terms of cost and time to implementation because we do recognize that it's a technology and business problem. We take care of the technology side so that the project can focus more on the business.

So you reduce the cost but you increase the flexibility. Changes are much more easily done within BIReady, is that correct?

Steve Aspris: That's correct. A good example of that is when a customer has an existing BI solution in place. When viewing a sample report, these customers often say, “Well the report that you're showing me is very good, but can I have this change and that change?” The traditional response for that person would be, “Well, I'll see you again in four weeks because those one or two changes to those attributes are going to take us a long time. We have to reinvent much of the system and dependencies that go all the way back to source systems.” However, with BIReady, since we're automating the process, we can now say to them, “Well why don’t you grab a coffee while I rework those changes, and I'll show you the results in a few minutes.”

Well that's excellent. Now Gertjan, you recently made some very interesting product announcements, and one of the reasons for this Spotlight interview is your ability to interface with CA ERwin. Can you provide our readers with a little more detail about that announcement and what it means to ERwin users?

Gertjan Vlug: Yes, I'm happy to do that. Since BIReady is a model-driven software tool, we are very interested in vendors who allow customers to create business models. That is why we started to talk with CA, their distributors, and their customers. We also work with Sybase as their PowerDesigner product also creates business models. By developing business models, organizations can formalize their business processes and business rules, and also define the KPIs. That is a good place for them to start development of applications.

But it's still a theoretical model. It's on paper, and it doesn't really help in the implementation phase. That, of course, is where BIReady comes in. With just a few clicks, we input the model and create the data warehouse. That combination, of course, is critical. Because the tool is taking care of all the data warehouse IT complexity, the business is far more self-supporting in creating prototypes. They engage better with IT now that the business people are taking more active ownership of the solution, and shared understanding increases. It makes life easier.

You're really solving an issue that has been a major disconnect for most enterprises. Many times, we create a model and then when we have it implemented, there's an interpretation by the developer as to how it should be done within the data warehouse. By automating it, you're actually eliminating that concern.

Gertjan Vlug: Yes, there are two important points here. First of all, since BIReady is so flexible that during implementation if you find out that your model might not be 100% correct, you can easily change the model. BIReady will apply the changes on the fly to your data warehouse environment. So it is not a real problem if you find something that needs to be changed. That's one aspect.

Secondly, with BIReady you only have to model your business. If you look at traditional modeling people, they model the business, then they start modeling the source systems, then they model the data warehouse, followed by the dimensional star schemas and so on. They make a lot of interdependent models. That is not necessary anymore with BIReady. The only thing that you have to model is your business as it is today. Once you have that, BIReady will transform that into a third normal form data warehouse and its star-schema access layers. The modeling itself is much easier than it used to be. The different variants of your business model are not necessary anymore. Therefore, the risk of critical requirements being lost in translation between models is eliminated.

Well that's great. Another relationship you have is with Teradata. When I look at data warehousing, the big data warehousing vendors are Teradata, IBM and Oracle, and you have a relationship with Teradata. Can you elaborate on that?

Gertjan Vlug: Yes, of course. We have a relationship with all of those you mentioned and also Microsoft. Teradata created the first real appliance for data warehousing. They are very strategic, and they have a huge installed base. That was, of course, the reason for getting Teradata supported by BIReady. We worked with their R&D people, and we made it happen. Then we talked to their experts about how it should be optimized for the Teradata platform in terms of performance. BIReady does it exactly the way Teradata advises their customers to implement data warehousing, the structures and the models and so on. It is a very good fit because both Teradata and BIReady are focused on data warehousing. IBM has a lot of other special focus areas and the same for Oracle and Microsoft. Teradata and BIReady both have a 100% data warehousing focus.

Also, Teradata recognizes the value of BIReady. Their customers can get the return on investment from their appliance much faster with BIReady – much sooner than if they had to painstakingly handcraft the whole solution. A lot of attention is turning to Teradata's LDMs, their logical data models. They have created authoritative models for different market segments, and it happens that they maintain and deliver those models within an ERwin environment. If people already have an ERwin model, BIReady can import it; and with a few clicks, their whole Teradata implementation is in place.

One of the latest products Teradata is developing is what they call Teradata Mapping Manager. So if you use that tool for mapping to your different sources in combination with the BIReady data warehouse environment, then the whole process is almost automated and that, of course, is a great situation for our combined customers. We are going to market together to find the right customers to make use of these advantages.

Gertjan, as long as we're talking about the advantages for customers, could you give us some specific customer examples of the time and cost savings that they’ve seen with BIReady?

Gertjan Vlug: Yes, I'm happy to do that. By automating implementations, you save time, and money, and the result is more standardized, optimized, and flexible. That is what everybody achieves with BIReady. A good example is one of our Dutch customers. They already had an offering from a system implementation partner to build a data warehouse. Then BIReady came in and created a prototype with that customer to figure out what the IT savings would be in their environment.

After Deloitte conducted an audit for that customer, we found out that they saved over 50% by using BIReady. On just the data warehousing portion of the product, it probably approached 90% savings. Of course, even with BIReady, they still had to understand their business and requirements – but the technical coding and modeling were done by BIReady, and that made it a success for this customer.

Steve Aspris: I’d like to add just one other point here. If we take a look at software development, we can understand the advantages of using Agile methods where instead of creating the entire application in its entirety before releasing it, it's the ability to create the most needed functionality first and then continue to build the application while the current release is actually bringing return on investment to the business. And this has been very, very difficult to achieve in data warehousing because of it being so expensive. When a company has to spend to build a data warehouse by hand, they face months of construction without being able to release anything. But now with the cost and time issue solved, we can get Agile with data warehouses. We can think big, but start small. We can prototype, test, and we can release certain reports into the organization and gain return on investment for them while changing and adapting quickly. It's only now with a real data warehouse management system that we can achieve truly Agile business intelligence.

Very good. Gertjan, do you have another example?

Gertjan Vlug: An insurance company saved over $800,000 a year and determined that the accuracy of their reports was dramatically better than their previous approach using ETL tools and traditional modeling methods. In addition, they were able to eliminate an expensive ETL tool that had been in place for years. The savings from that license greatly outweighed their investment in BIReady.

While cost savings are a major benefit, the real advantage is in the time to market of your business intelligence system. With BIReady that's where you get your efficiency and your competitive advantage.

Thank you Steve and Gertjan. I appreciate all the information you've provided to our BeyeNETWORK audience about the BIReady model-driven approach to data warehousing.

SOURCE: Avoiding the Pitfalls of Data Warehousing: A Spotlight Q&A with Gertjan Vlug and Steve Aspris of BIReady

  • Ron PowellRon Powell
    Ron is an independent analyst, consultant and editorial expert with extensive knowledge and experience in business intelligence, big data, analytics and data warehousing. Currently president of Powell Interactive Media, which specializes in consulting and podcast services, he is also Executive Producer of The World Transformed Fast Forward series. In 2004, Ron founded the BeyeNETWORK, which was acquired by Tech Target in 2010.  Prior to the founding of the BeyeNETWORK, Ron was cofounder, publisher and editorial director of DM Review (now Information Management). He maintains an expert channel and blog on the BeyeNETWORK and may be contacted by email at rpowell@powellinteractivemedia.com. 

    More articles and Ron's blog can be found in his BeyeNETWORK expert channel.

Recent articles by Ron Powell



 

Comments

Want to post a comment? Login or become a member today!

Be the first to comment!