Possible Futures for Enterprise Architecture

enterprise architecture predictions

As I envision where enterprise architecture will go in the future, parallel activities in meteorology come to mind: how similar the enterprise architecture role is to meteorology and how predicting the future for our occupation is much like predicting the weather. In both meteorology and enterprise architecture, we leverage data, events, historic trends, and patterns to build the models to base decisions on and to realize our expected outcome. Enterprise architects often forget our constituents expect us to answer certain questions with imperfect knowledge. While everyone would like to do their job accurately, imperfect knowledge is the grim reality. I’ll address three topics I strongly believe will evolve to fundamentally impact the enterprise architecture role in the next few years:

  1. Enterprise architecture and related standards will merge to become mainstream.
  2. Enterprise architecture will focus more on business and service architecture, and less about IT.
  3. EA tooling will merge and evolve into seamless enterprise management practices.

ENTERPRISE ARCHITECTURE AND RELATED STANDARDS WILL MERGE TO BECOME MAINSTREAM

The outlook for standards in enterprise architecture is promising. There are now many standards and frameworks in play to address slightly different needs and preferences including Zachman, FEA, DODAF, FEAF, PEAF, and TOGAF, to name just a few. TOGAF pulled to the front most recently in commercial and some governments, gaining adoption as the leading standard today. Currently at Version 9.1 with more than 26,000 certified architects worldwide, TOGAF adoption continues to grow as a truly global standard with a mature ecosystem of supporting education, practitioners, and tools. Given TOGAF’s trends, we can expect this evolution to continue, even slow down, given the large number of stakeholders involved today.

A key area to watch is where TOGAF collides with related or overlapping standards. For example, ITIL addresses overlapping concept with service strategy, design, and transition. ITIL was born from an IT operations point of view, evolving toward business and strategy through ongoing revisions. COBIT, a popular IT governance and controls framework, continues to evolve, too, overlapping with both ITIL and TOGAF in key governance topics. Consider the overlaps in ARB (Architectural Review Board) and CAB (Change Advisory Board), for example. It’s likely that overlaps between these three standards will continue to evolve and increase, forcing a more comprehensive standard to emerge at some point in the future. At a minimum, better references to one another are needed.

Another area for standards evolution that we can safely predict is the incorporation of business architecture into the standards bodies and frameworks. IT continues to be commoditized by cloud, IaaS, PaaS, and SaaS initiatives that effectively abstract technology implementation from use. Our focus must now orient toward business architecture as a main concern to drive needs for services and technology. Since business funds investments and business decision makers are ultimately responsible for technology investments, it makes sense to close the gaps wherever possible. It astounds me that most IT organizations have little to no business architecture in play, effectively making most current and future technology investment impossible to trace to business outcomes. For this reason, business architecture must become a focus area for standards evolution and EA to be truly relevant to the organization. Currently, both the OMG and Open Group are feverishly defining new business architecture standards. We can expect new standards to emerge that will bring EAs relevance to the business one step closer in the very near future.

ENTERPRISE ARCHITECTURE WILL FOCUS MORE ON BUSINESS AND SERVICE ARCHITECTURE, AND LESS ABOUT IT

This prediction is based on numerous observations over the past few years from analysts reporting through firsthand experiences with a handful of customers leading business transformations today. I imagine a world where an organization can be expressed as a series of holistic models such as the Business Motivation Model. EAs can make changes to the model then test the outcomes and impacts in analytics and what-if scenarios. A complete understanding of impacts and alternatives will be necessary. Today, these exercises are rather disparate. I would characterize most of them as fire drills with all hands on deck to plan or execute.

Another area that will evolve is services modeling, supporting a better modeling scenario for connecting business and resources in a more abstract and contractual perspective. A full-service oriented management paradigm is the critical step to enable an agile organization, similar to how SOA fundamentally changed development to be more agile and configurable for application development. Services defined at the business, applications, information, and infrastructure levels and that express their capabilities in consumer terms can make this possible. Some services can continue to be sourced within an organization, while others will need to be sourced externally as decided upon by core vs. contextual analysis (a key driver).

Service portfolio and catalog management will become the most critical enabler for the business architecture, providing necessary viewpoints to determine the services necessary to achieve a desired business outcome for the business model now and the future. It will become increasingly important to reconfigure businesses quickly, employing the right services that deliver expected business capabilities. Business competition is evolving and changing more rapidly today than ever, and it’s not going to slow down. Entirely new markets are emerging from nowhere, threatening traditional market players in every industry. Time to market, quick feedback, and failing or succeeding fast trump lengthy investments in the information age. Service-oriented management abstracts implementation from needs, allowing a rapid reconfiguration without the baggage of understanding service internals. Today’s immature service modeling standards will need to be improved.

EA TOOLING WILL MERGE AND EVOLVE INTO SEAMLESS ENTERPRISE MANAGEMENT PRACTICES

Tools used by enterprise architects have evolved from largely niche players in the market to being acquired by larger software portfolio players with the most aggressive investments from IBM and Software AG, each of which has started to integrate EA tooling into its larger tools ecosystems. Software AG purchased IDS Scheer in 2009 and Alfabet in 2013. Looking at broader trends, it’s clear that integrated EA planning to execution and delivery is occurring in the tools space.

Continuing the trend, we can expect modeling and analysis activities to seamlessly integrate business design through operational implementation with fewer steps and delays, approaching an almost real-time return on investment from a business outcome point of view. While fully integrated scenarios remain elusive today, abilities to deliver this story is just around the corner.

In the past few years, the TOSCA standard has evolved to address the use cases above DMTF with vendors such as IBM and HP participating. They have recently demonstrated their ability to deploy the same solution from a model in their respective cloud platforms. Enterprise architecture and downstream tools are closing gaps to front-end these traditionally disparate steps into a seamless value stream. There are many efforts breaking down traditional silos and accelerating activities, such as in agile development methodologies or DevOps for example.

One last area to watch is how “Big Data” plays a role in enterprise architecture tools. We are only beginning to grasp how to leverage large, unstructured data sets in traditional jobs such as marketing and business intelligence. I envision a time where modeling the enterprise can take advantage of unstructured data, too, such as archived project documents, older Visio diagrams, requirement repositories, and even executive MBOs to understand the enterprise’s current and future direction. Possible Futures for Enterprise Architecture A&G Doing so removes painstaking activities to gather and map information into useful modeling notation. Since every organization has a deeply rooted unique lingua franca of its business, preserving context using inference engines would accelerate enterprise architect tasks.

Expect mobile and social evolutions to impact enterprise architecture, too, most likely in feedback, collaboration, and what-if planning exercises. One large financial firm I worked with has already created an iPad application used by IT executives to view enterprise architecture, provide feedback, and collaborate. While it’s rare to see this level of maturity, it exists and provides an indication of things to come.

IN SUMMARY

Enterprise architecture can expect to be influenced by the same forces influencing any practice today. Standards will continue to evolve, merge, and become more commonplace as we are educated on cross-job requirements. As our practice changes under our feet, I encourage you to get involved, help our practice standards evolve, and step outside your comfort zone. Take a weatherman approach to EA—provide forecasts and models that produce guidance for your leadership. Don’t expose models directly; focus more on the desired outcomes your leadership is looking for and produce recommendations that can help ensure a successful business outcome. Enterprise architecture has real potential to become the integral component for planning to delivery for any organization. To do so, we must each get involved, embrace evolution, and aspire to close gaps between activities. It’s an exciting time to be an enterprise architecture. Enjoy the ride!

Avatar
About Mark Bodman 1 Article
Mark Bodman is an enterprise architect at HP Software.