|At General Motors enterprise architecture is a dynamic, living process dependent upon integration with related enterprise process disciplines.
General Motors' chief architect Richard Taggart says. "Enterprise architecture is quite honestly a fairly new practice. The whole idea of intellectual capital - there's not a lot of work done there either. The two fields really haven't discovered each other yet, but I do believe it's inevitable because what they want to do in intellectual capital management is to find ways of capturing the knowledge that's in people's heads and the resources that work in a company, and to leverage those resources better around the company. And what architecture allows you to do is to make explicit the implicit."
Taggart has been managing the development, integration and rationalization of enterprise architecture across GM's business, infrastructure, application development efforts and operational architecture functions. He has also overseen the creation and management of the corporation's Information Technology Standards, assuring that the life cycle of the technology investment is aligned with GM's strategic business goals.
Taggart works for the chief technology officer, who heads one of the few global organizations within GM, and is responsible for the global practices of standards and architecture. He works closely with subsidiary organizations which he says gives him a "very nice horizontal view of the company, both in terms of all of the standards that we want to have within IT, as well as trying to move along the practice of architecture."
Taggart has developed an Intellectual Capital Maturity Model building on EA standards and principles. Last year Taggart co-founded the Enterprise Architecture Interest Group (EAIG), whose members include Volkswagen, DaimlerChrysler, Booz Allen Hamilton, Oakland University, Sandia National Laboratories and the Zachman Institute for Framework Advancement, a group that promotes the Zachman Framework as a starting point for describing enterprise systems. EAIG was formed to fill a void by standardizing EA definitions and practice and to help companies develop and share tips on building well-structured IT systems, study ways to measure the benefits of formalizing enterprise architecture and develop value models for use by EAIG members and their organizations.
The group has released its first deliverables, a set of 12 meta-models intended to be used as architectural building blocks to create any desired model of any business. This first deliverable, which Taggart says essentially amounts to a business architecture, was built around the Zachman Framework. "These 12 models represent a comprehensive set of reusable building blocks to represent any organization or any enterprise," Taggart says.
To make all of its intangibles explicit, GM has developed the Intellectual Capital Maturity Model, and will shortly begin testing the effectiveness of those basic building blocks in capturing intellectual capital. The model is the outcome of extensive work by Taggart in exploring fundamental techniques for modeling a business, and can be used to categorize tangible or intangible assets - processes, components, solutions and organization - according to whether they are unrecognized, recognized, chaotic, organized, assetized, managed, valued or monetized.
GM also has developed a concept called 'Business on a Page,' which lets it show the goals, major functions and major pieces of information the organization manages and maintains to provide performance functions on a single chart, allowing businesspeople to see their business at a glance, take ownership of it and be aligned around it. Taggart says these are proving very popular with businesspeople, and particularly with HR, who see Business on a Page as a great way to begin training for new employees.
Creating enterprise architectural blueprints can also help companies cut costs with outsourcing, Taggart says. "It helps you understand what can or can't, and should or shouldn't, be outsourced."
Taggart has already used the business architecture to create an extensive model of all of the functions the GM enterprise performs. The model started at very high level and then drilled down to much greater detail - a thousand-plus entries - in key parts of the organization such as manufacturing and engineering. He then developed a technique to map GM's entire application base against those functions, giving it a "very clean" understanding of what applications are supporting which processes at GM. He has also developed techniques for mapping organizations according to functions, competencies and systems.
Taggart says the model makes it very easy visually to recognize opportunities for eliminating redundant applications, and areas where the organization lacks applications and needs new initiatives. For instance, it let GM identify an opportunity for a new call center application for its OnStar subsidiary, GM's successful telematic system for safety, security and online road service assistance.
"We did some work where we created a business model of the OnStar business. We clearly saw some important initiatives that were missing from our portfolio, which have now been addressed, where we saw that there were some functions that were rated as being high strategically. In other words, these functions supported the goals of the organization - but we hadn't adequately mechanized them. We saw major gaps.
Taggart also used similar techniques, based around those business models, to complete a worldwide SAP analysis - for CIO Ralph Szygenda - delineating how GM was using SAP across the globe, and pointing to opportunities to develop an initiative for further leveraging of SAP in a way that would be highly beneficial to the company.
"We found numerous ways that over the course of the next 18 months, we would be able to better leverage SAP inside of GM, and in fact we are doing that," he says. (One of CEO Rick Wagoner's objectives is to introduce common processes in the business. He has called "going global" with processes a major strategic objective of the corporation.) "We used this technique as an analysis technique to identify which processes needed to be further globalized, and we identified those, and now we're in the process of remediating," Taggart says.
Taggart observes that by using what is basically a very simple but powerful analytic technique GM has discovered opportunities for cost reductions, additional systems initiative opportunities and clear gaps in its portfolio that have led to new initiatives to fill in those gaps. For example, it discovered a way to leverage SAP for indirect material procurement on a global basis.
"It's primarily an analysis tool. It doesn't do the work for you - it's primarily an analysis technique that can show you where your gaps are," he says. "The real benefit of having business models is that you can create a model of where your business is today, and you can create a model of where your business needs to be in a year, in two years or five years - whatever time frame you want to pick - and then you can look for all the gaps. You have a way of analyzing how are you going to get to where it is you want to go."
The technique also gets to the very heart of business goal alignment by explicitly expressing the goals of the business and by providing, in architecture, a way to make strategy happen. "It gives you your road map to take your strategic initiatives and to show point by point the things we need to do to move towards where it is that we want to go," says Taggart. "It's great to have strat plans - strat plans are wonderful tools, but they sit on somebody's shelf. How do you make them actionable? Architecture can turn a goal into an actionable plan, if it's done well."
However useful the methods and techniques, Taggart says developing an enterprise architecture for a company the size of GM - given that the sheer scale of the company is breathtaking - is a daunting task.
"GM is 350,000 employees; it's in 200 countries plus around the world. Our Web sites, I've been told, reach 3.5 billion people, so the scale of the company is a bit beyond belief. And just simply dealing with that scale represents all kinds of problems and issues about getting the right information to the right people, internal alignment - all those kinds of things are huge challenges because of the scale of the company.
"Most of the time, we deal with it simply by getting one convert on board at a time. Sometimes you have a corporate mandate that you can leverage, and I have when that's been clear. Other times it's really a matter of just educating people and getting people on board one by one and just moving the ball forward," Taggart says.
Since architecture cannot afford to remain static in a dynamic company, keeping the architecture up to date is a never-ending task. "Architecture must always be refreshed and evergreen, since any architecture represents a particular point in time," Taggart says. As soon as any "As Is" architecture - designed to reflect the current state of the company - is finished, the work of updating it must begin, he says. That means the architect must make a commitment to continue to update and work on all of the architecture's elements, and assign diverse people the various responsibilities of managing and upgrading it. If you do not, the information goes out of date very quickly and loses its worth.
"That's why these reusable models and standards in this base are so important," he says. "If we can develop a set of standards in this industry around architecture, we can now have certified professionals. We get people who are certified to produce a kind of deliverable, to maintain a kind of deliverable. And I'll have a much easier time evergreening all this work. Without a sufficient set of standards and a certification program, certification bodies and so on, the real danger is that you wind up redoing work over and over again as opposed to reusing it - meaning a missed opportunity to save money."
(extract of an article by Sue Bushell, CIO magazine, October 20, 2004. Copyright 2004 IDG Communications)