There is a shift underway with master data management (MDM) that can't be ignored. It is no longer good enough to master domains in a silo and think of MDM as an integration tool. First-generation implementations have provided success to companies seeking to manage duplication, establishing a master definition, and consolidating data into a data warehouse. All good things. However, as organizations embrace federated environments and put big data architectures into wider use, these built-for-purpose MDM implementations are too narrowly focused and at times as rigid as the traditional data management platforms they support.
Yet, it doesn't have to be that way. By nature, MDM is meant to provide flexibility and elasticity to managing both single and multiple master domains. First, MDM has to be redefined from a data integration tool to a data modeling tool. Then, MDM is better aligned to business patterns and information needs, as it is designed by business context.
Enter The Golden Profile
When the business wants to put master data to use it is about how to have a view of a domain. The business doesn't think in terms of records, it thinks about using the data to improve customer relationships, grow the business, improve processes, or any host of other business tasks and objectives. A golden profile fits this need by providing the definition and framework that flexes to deliver master data based on context. It can do so because it is driven by data relationships.
The InfoWorld/Forrester Enterprise Architecture Awards recognize excellent EA programs — ones that due to their business focus, and strategic yet pragmatic orientation, provide sustained value to their business. I caught up with two of our 2012 winners to find out what they have been doing in the year since their award submission. I was specifically interested in hearing:
Have there been changes to business strategy or IT strategy since one year ago that they’ve had to respond to?
What would they say has been their greatest accomplishment over the past year?
The priorities for their EA programs today — changes in the scope, mission, or organization?
What would they say is a key learning of their EA program, or the larger IT organization about making EA effective?
I met with a group of clients recently on the evolution of data management and big data. One retailer asked, “Are you seeing the business going to external sources to do Big Data?”
My first reaction was, “NO!” Yet, as I thought about it more and went back to my own roots as an analyst, the answer is most likely, “YES!”
Ignoring nomenclature, the reality is that the business is not only going to external sources for big data, but they have been doing it for years. Think about it; organizations that have considered data a strategic tool have invested heavily in big data going back to when mainframes came into vogue. More recently, banking, retail, consumer packaged goods, and logistics have marquis case studies on what sophisticated data use can do.
Before Hadoop, before massive parallel processing, where did the business turn? Many have had relationships with market research organizations, consultancies, and agencies to get them the sophisticated analysis that they need.
Think about the fact, too, that at the beginning of social media, it was PR agencies that developed the first big data analysis and visualization of Twitter, LinkedIn, and Facebook influence. In a past life, I worked at ComScore Networks, an aggregator and market research firm analyzing and trending online behavior. When I joined, they had the largest and fastest growing private cloud to collect web traffic globally. Now, that was big data.
Today, the data paints a split picture. When surveying IT across various surveys, social media and online analysis is a small percentage of business intelligence and analytics that is supported. However, when we look to the marketing and strategy clients at Forrester, there is a completely opposite picture.
As the analyst covering all things emerging information technology, I spend a bit of time watching web and social feeds looking for interesting and potentially disruptive stuff. Fortunately, it’s a good time for me to be doing this, as there are all kinds of things going on. I’ve decided to pass some of the best on to my readers in periodic “What’s Cooking” posts.
Digital currency will turn retail and financial services on its head, eventually. Digital currency fascinates me, especially the enigmatic Bitcoin creator and its so far unbreakable code*. Also the way you have to mine for more coins is very interesting. Whether or not Bitcoin succeeds as the de facto standard, I think digital currency is inevitable and the more firms that accept is, the crazier things will get. Check out The Antisocial Network of Bitcoins.
Recently, I interviewed a half dozen top service design agencies to better understand how they work with enterprise architects and business architects inside the client firms they serve. All of the agencies I interviewed focus on helping their clients transform customer experience and introduce new products and services. I wanted to interview these agencies because they represent the tip of the spear when it comes to introducing new innovation inside of companies looking to take advantage of disruptive drivers - both competitive and digital – and rethink their business models.
I asked each agency for examples of how they worked with their clients’ enterprise architects and business architects when introducing new innovation. When I posed this question to each agency, I could hear crickets chirping in the background. In short, they all indicated – in as nice terms as possible – that they try to avoid the IT organization in general, and had no contact with specific enterprise architects or business architects.
For me these interviews painted a picture of a school yard where team captains are picking players for kickball, and a small group of kids were being left on the sidelines, not picked for the team. Using this analogy, the business – in many cases the CMO and CXO leaders – are the team captains. And enterprise architecture, including business architects and process architects, are the kids being left on the sidelines.
I heard a great analogy from a client recently; buying new technology is like buying a new car - there are a lot of different strategies. Some people want a new car every couple of years and pay a premium to have it, some choose to lease so they get a new car every few years at a lower payment but they don’t own. Others buy new but plan to drive the wheels off their purchase. The problem is that IT wants to buy a nice reliable sedan and drive it for 200K miles, while some business units want to lease a SUV and others want a Ferrari. It’s an issue of misalignment, but in so many cases IT is not synching up with the business desire to innovate and differentiate with new technology.
Many architects and technology executives relate a cautious approach to introducing new, “bleeding edge” technology because they are in a very conservative business that doesn’t change that much. Ask them about the level of business investment in technology outside of IT, however, and they whistle. “Yup, that’s happening allot.”
Lots of organizations I speak with are undertaking rationalization and simplification of application portfolios - some successfully, some less so. I think that the key difference between successful and less successful rationalization is the order in which organizations go about it. Organizations that rationalize their applications successfully start with (1) business drivers that define the need for change, then (2) they define the characteristics of the application portfolio that will enable those business changes. They then (3) outline a program of phased activities that will deliver the portfolio changes and finally (4) undertake a series of projects that make the functional, technical or process changes. Less successful programs take a different sequence - it typically goes in a 4-2-1-3 sequence, starting with projects, then portfolio changes, which deliver some business change that are wrapped up in a program.
When I posted a blog on Don’t Establish Data Management Standards (it was also on Information Management's website as Data Management Standards are a Barrier) I expected some resistance. I mean, why post a blog and not have the courage to be provocative, right? However, I have to say I was surprised at the level of resistance. Although, I also have to point out that this blog was also one of the most syndicated and recommended I have had. I will assume that there is a bit of an agreement with it as well as I didn't see any qualifiers in tweets that I was completely crazy. Anyway, here are just a few dissenter comments:
“This article would be funny if it wasn't so sad...you can't do *anything* in IT (especially innovate) without standing on the shoulders of some standard.” – John O
“Show me data management without standards and good process to review and update them and I'll show you the mortgage crisis which developed during 2007.” – Jim F
“This article is alarmingly naive, detrimental, and counterproductive. Let me count the ways…” – Cynthia H
"No control leads to caos... I would be amused to watch the reaction of the ISO engineer while reading this article :)." - Eduardo G (I would too!)
After wiping the rotten tomatoes from my face from that, here are some points made that get to the nuance I was hoping to create a discussion on:
It's becoming pretty clear that the ability to analyze data is becoming one of the most important technology-based capabilities an enterprise can have. There's a lot of hype around about big data, and it's actually well-founded hype --- if that's not a contradiction (perhaps I should call it well-founded fanfare). In any event, our world is changing as organizations gain the ability to process formerly unheard-of amounts of data with formerly unheard-of speed. New, improved information processing capabilities are significantly changing science, where scientists in labs look for patterns in data rather than dream up hypothoses and run tests to prove them right or wrong. And, in similar ways, it's changing how businesses make decisions. I've been looking for evidence that enterprises are moving on improving their information management capabilities since we started doing our "State of EA" surveys in 2009, and the 2012 data finally shows that developing or expanding information architecture is finally EA's #1 priority (well, OK, it's tied for first place with developing or expanding business architecture).
Information Architecture Is Finally A #1 EA Priority
SAP launched its HANA in-memory computing platform in 2010. HANA is a converged analytics appliance. Three years later, SAP has officially launched Business Suite on HANA: globally in January and in China on March 19. SAP clients can now run mission-critical applications on the converged infrastructure for optimized performance. Personally, I would suggest calling this an example of converged applications, which in short refers to the business applications that are architected around the converged infrastructure for performance and simplicity.
I had several conversations with architects from the retail, logistics, and manufacturing industries, as well as Tom Kindermans, SAP’s senior vice president of applications for APJ, about these converged applications. I tend to believe that this is the next wave of application architecture, after mainframe, client/server, and browser/server. With the deployment of these converged infrastructure offerings and the evolution of the applications that run on top of them, it might change technical architectures across infrastructure, information, and applications, as well as the organizational structure of IT, the architecture, and the partner ecosystems. My assessment:
The definition of converged applications is blurry. The meaning of incorporating converged applications can vary quite a bit. Sometimes it means migrating an application from one server to the other; sometimes it means refactoring your networking and storage design for load balancing and disaster recovery; and sometimes eliminating an original performance bottleneck means that business challenges that had been lurking under the surface might emerge for you to resolve. It totally depends on your business goals.