Why Frameworks Are Killing Architecture

Let me start by contradicting my headline. I love frameworks. DODAF, FEAF, MODAF, TOGAF, Zachman etc. are all excellent tools for architecture teams to use when they are ready and when the team members are skilled enough to use them. Each brings different strengths and weaknesses, and can help an experienced team do very interesting things. But. sadly, the vast majority of enterprises approach frameworks as solutions to good practice as opposed to tools to support a people and experience focused engagement models. This approach has taken architects from innovators to documentation specialists and continues to destroy both our capability and credibility in the market place.

  1. Frameworks get in the way of innovation: The heart and soul of architecture success is digital transformation. Existing frameworks are deliverable and process focused and tend to take the focus off of transformation and put the emphasis on governance and IT operating stack. Take a look at http://www.infoworld.com/article/3121276/it-management/want-to-transform-your-business-ask-an-architect-to-lead.html. In this article, the author says what Iasa and I have been saying for years. Never let yourself be removed from innovation and digital transformation for business outcomes. If you do, your team will fail. The frameworks today are primarily about documenting the enterprise not leadership with a skills focus. For example, if I have my daughter memorize scrum, it doesn’t make her a good developer. This is roughly equivalent to what frameworks have done to the transformative architect. By focusing on TOGAF, they have become documentation specialists not innovators.
  2. Frameworks put the focus on deliverable instead of outcome: If you took a loved one to a medical professional to fix a broken arm and they simply gave you a document which described a broken arm and a healthy arm, you would run screaming to someone qualified to solve your real problem. You go to the doctor to get that arm FIXED. Our current framework focus does exactly that with architecture. Instead of being liable for business outcomes, architects are, at best, responsible for some esoteric deliverable, which in practice does not get used. In many cases architects are not even responsible for the deliverable, but simply sit around in a big group and ‘review’, ‘govern’ or ‘approve’ designs that other people do. Take a serious look at TOGAF. Is it focused on the architect’s role and responsibility in outcomes? Or is it a generic set of steps and deliverables? As I said, frameworks are great, deliverables and process steps are useful, as long as the role is accountable for outcomes.
  3. Frameworks create poorly skilled architects: The only common element to architect job descriptions en masse has become framework specific. For example, putting TOGAF or Zachman on an architect job description has put the focus for architects on knowing those frameworks not practicing effective architecture skills. This is roughly the equivalent of teaching someone how to use a nail gun and calling them a carpenter. Or teaching them the legal shapes on a blueprint and calling them a building architect.
  4. Frameworks make hiring architects more difficult: Quality is about skill and experience, not framework utilization and memorization. It is true that a great architect will be more successful with a rigorous practice, process and toolset. But a poor architect will be a poor architect regardless of framework. The ONLY meaningful way to refocus our practitioners is to begin to require experiential certifications before hiring. This puts the onus on the individual to develop their skills and experience and moves away from ‘tool’ and ‘process’ training/certification.
  5. Frameworks allow architects to keep ‘killing patients’: Our field is difficult, as are all professions where depth of knowledge and breadth of experience so drastically differentiate the good from the bad practitioners. Many, many years ago, medical doctors had roughly the success rate we do because they understood so little of the art and science of their practice. By success, I mean that long ago, doctors regularly killed their patients. Our current practitioners have the same problem. They will regularly ‘kill the patient’; cause project, program or outcome to fail based on a lack of experience and knowledge and a history of success of the architect themselves. In addition, as a field we lack evidence gathering capabilities to advance our rigorous understanding of what success really means. For example, if I ask 4 different architects whether a particular architecture would be successful and how they would measure the success, I will get 4 different responses (I have done this experiment regularly for many years). We, in fact, have very little evidence currently to describe what a ‘good’ architecture actually is.

I often get chided by my friends that I will call attention to a problem instead of a solution. I admit I have a distinctive ‘kids get off my lawn’ kind of writing style. So I will attempt to provide a simple and positive answer to the framework issue.

  1. Move to only working with individuals who a have achieved an experience based certification in architecture: This is the easiest, cheapest and most long-lasting solution to skills and excellence readiness for architects in the industry. Think of it in the following way. Would you work with a non-licensed physician regardless of how cheap you could acquire another’s services? No of course not. If you change your job descriptions and service contracts to require board certified architects, the architects who want jobs will achieve them and you will be able to hire based on culture fit and specific business elements (industry, capability, technology, etc) without worrying that the doctor will kill the patient. Of course, I realize that the supply of these individuals is low, so instead of moving these to required immediately, do so over the next 18 mo. Just so you know I am not being self-serving, both The Open Group and Iasa provide acceptable board certifications.
  2. Optionally pay for the personal development roadmap of your current people, not your future staff: As an extension of point one, you may choose to pay for the development of your current people along that 18 mo roadmap. But refuse to do so for senior hires of the future. (Honestly, though, it would send the right message to the professionals for them to pay for their own personal development, as is expected of ALL other professionals in every industry.) Explain to your service providers that they have 18 mo (or less) to provide this quality of architect professional or you will be utilizing another provider who will. While this may seem draconian and you may be concerned about supply of professionals it takes only a critical mass of companies doing this to achieve the momentum we need.
  3. Measure your architects based on business outcomes: I have always said, what would your architects make if you paid them only on commission. If they are not willing or able to give you that amount, hire new architects.
  4. Start measuring outcomes related to design decisions and share results: Are the patients dying? Are the buildings falling over? When, how, what happened? As a field we must begin to understand the forces which shape solutions and what works and what doesn’t in a much more scientific and rigorous way. When is cloud the answer and when not? When is IoT a business initiative and when is it more IT focused? Does an agile approach make sense, why and why not? What is the data that is driving our profession to further excellence? Are we simply hype-makers and product evaluators? Or do we have a larger ethical responsibility to our solutions?

There is as much nuance and complexity to this situation as there are common and simple solutions. But ask yourself one fundamental question (which will sound a little out there). Is architecture more complicated than medicine? If you answered ‘of course not’ the same way I did, then why are we focused on good ‘process’ and ‘tools’ instead of focusing on creating and sustaining the best architects in the world? The doctors did it, why can’t we?

About the Author:

Paul Preiss is the CEO and Founder of the Iasa, one of the largest Enterprise and IT architect associations in the world. Through his time at Iasa, Paul has taken the association from a single user group in Austin Tx to an international organization with chapters in over 25 countries. Paul’s vision is a unified architecture profession with effective education, credentials and ethics which fully supports corporate strategy and delivery. He is a tireless advocate for the field and speaks on topics ranging form architecture ethics to best the best setup and structure for architecture teams. Paul has spoken at hundreds of events as well as held conferences and training for architects all over the world. He is an expert software and enterprise architect in practice and continues to work with companies on optimizing their technology strategy. Prior to developing Iasa, Paul was the chief architect for Dell Pan Asia where he helped to integrate the technology strategy across 14 countries. He also served as the chief architect for the Sears point of sale replacement in North America consisting of 2000 stores and thousands of suppliers as well as the chief architect for a digital asset management firm, Ancept.

One Comment

  1. Loan Kenney November 22, 2016 at 2:12 pm - Reply

    Thanks for sharing your thoughts Paul. I wholeheartedly agree with you. Enterprise architects are becoming frameworks evangelicals as supposed to architectural practitioners.

Leave A Comment