Updated Model of Enterprise Architecture Function Value

How Technical Debt is Created

Measuring Technical Debt is not a single metric.  There are many factors that come into play, most are not visible to the leadership or within the product development lifecycle unless these metrics are specifically called out for in the process.  Added to this Technical Debt Visibility issue is the addition of newer development models such as Agile.  These models foster speed and agility, but if implemented by the unaware remove visibility to the Debt created.   Add market, customer, and management pressure for faster delivery and one can create condition to take shortcuts rationalizing that once in production the team will go back and correct. 

However, rarely does this happen.  The results are the team is then rushed to the next project and next and so on.  Addressing issues is left to the helpdesk, operations team or escalated back to software development team to patch quickly to get the application back online.  Over time this can create a bowl of spaghetti code, thus increasing one aspect of Technical Debt.

The problem can and often is pushed down to developers. “We have a code quality problem” Sound familiar?  When the truth is much broader than that.  Technical Debt is a Process Quality Issue. 

Often the issues start at the program planning and requirement phase.  We all have heard the manager’s rule of thumb of 80/20 or the like.  If you spend 20 hours gather requirements it will take 80 hours to code.  Thus, with a deadline management often pushes to finish the requirements and design phase quickly.  This is where the first bit of debt is created. 

Hidden in that rush a complete understand of how the software is to function and the context it must operate in now and in the future is missed.  Thus, alternatives are not explored in the rush to create utility, warrantee is often missed.  In other industries and disciplines these lessons have been had learned.  The cost of failure can be catastrophic or financially devastating[1] at best.

This quality issue is compounded with each step in the lifecycle till operationally all that is left is application utility for the immediate, leaving the minimal warrantee and Technical Debt such causes to grow hidden over time till a situation or scenario causes it to explode.  

---

[1] AA, Southwest, and other Software failures that have devastated enterprises.          

Enterprise Architecture Value Chain

I’ve been asked many times what the value of Enterprise Architecture is. Most often the answers from everyone are very vague or have a lot of hand waving. This month as I review a lot of my past several decades of work, I’ve started to notice some interesting patterns. So, as I continue with my Establishing an Enterprise Architecture Practice Book and White Paper on Technical Debt, I’ve decided to work on another White Paper A model to Measure EA Function Value. Above is an early ArchiMate model of the Value Chain.

Enterprise Design and Transformation

Pondering the initial model in “Business and Dynamic Change”, this morning led to this 1st draft of Enterprise Design and Transformation Capability Model.  I had been watching a video from Productivity Games on YouTube (Well worth subscribing) on Performance State and Flow.  Suddenly I was in the state and this 1st draft on the next level of designing and transforming enterprises appeared to me.  This was looks like the framework I was trying to abstract from the process flow I created earlier.  The first chapter to the Structure in Threes (follow-up to Chapter in “Business and Dynamic Change” is on the agenda for this month.  

Investigating developing some instructional courses around the book to transfer skills to other EA.  Looking for the right platform for such now.

 

CEO’s Transformation Nightmare

This week I had an interesting time creating a detailed scenario and storyboard for circumstances enterprises face today.  It’s a mashup of several brainstorming techniques I learned while in a Think-tank decades ago and from time to time get ask to participate in again due to the insights and opportunities it creates. With that context here is a fragment of the fictional narrative I came up with. 

 

One of the chief executives in a rather large firm called me up in the middle of the night; this has really happened in the past, but not late at night.  He was concerned with their recent transformation efforts.  The CIO, CFO, and him had agreed to a course of action to transform the business by moving to the cloud.  The logic being big consulting firms, analysts, trade press, and vendors all were touting going to the cloud, so they should also.  After several months of preparation, execution, and significant expenditure of funds the majority of the enterprise was now “In the cloud”.  The CIO was happy, the CFO was happy, all seemed right but something was bothering him.  It was only a couple of months to the annual stockholders meeting and the board of directors meeting the week past left him with an uneasy feeling.  One of the questions during that meeting hung in the air like the feeling you get from oppressive heat and humidity:

 “You’ve spent X dollars this year decommissioning the data center just several years ago you argued we needed and an additional sum migrating our data and applications to the cloud.  What will the stockholders think about what they got for the money?  Sales and margins have not changed significantly from last year.  Our net profit is lower due to the additional expense.  While our IT expenses in the future will be more elastic changing from CAPEX to OPEX.  Are we going to be able to meet Wall street expectations and dividends our shareholders were expecting from completing the datacenter build out we’ve been touting up until the beginning of the year?  How is this going to grow the company?

 The discussion we had could have been the general business version of the book “The Goal”.  James relayed the discussion with the board and his concerns to me.  As in the book, I asked several questions knowing the answer already, but to have him consider his actions from a systems approach.  This was difficult for him so I reframed the questions into a more concrete example.  He had already read “The Goal” as captured its lessons when he was the VP of Manufacturing years ago.  So I asked if your IT Capability was a CNC machine on your floor would you have made the same decisions?  What?  He exclaimed.  Would you sell your CNC machine that you just bought two years ago and lease a new one today?  A few minutes of thought.  “Well no”  So I asked why then sell you “information machine” and lease a new one?  “Well it changes our costs structure from CAPEX to OPEX and it provides us a more elastic capability.”  Oh I see, so you’re planning to lay people off in the next few months and preparing to be acquired?  Are you about to ask about how to play the M&A game?  “NO! I had nothing like that in mind”  Well from the outside of your executive staff room, that’s what I’d assume is being prepared for.  “No, we’re preparing to utilize the cloud’s capabilities to expand our business and grow”  I ask one final question. How?  I could see his heart immediately sink.  This was what his subconscious was telling him about the board and upcoming shareholder’s meeting.  He’d gotten caught up around the hype and focused on the tool rather than its use.

 This small little nightmare is on the verge of happening all over the landscape as I watch “Cloud Transformations” occur.  If your about to “transform to the cloud” what you should be asking as the chief strategist is “How am I going to exploit this capability to improve the business?  Not the cost structure unless your cost structure makes you uncompetitive”       

Cloud and AI Adoption Maturity and Opinion

Developed in 1989 I observed that design disciplines such as Architecture, Mechanical and Electrical Engineering, Information Architecture, etc. followed a unique and recurring pattern that was intimately linked to technology.   Awareness of the technology occurred with innovators and early adopters exploiting its new characteristics first. As the technology’s performance attributes became better known application techniques developed through a maturity cycle becoming more structured. This body of knowledge eventually become ubiquitous in the discipline and the technology application common practice. Eventually, the discipline and technology application pass through industrialization stages to where practitioners debate about the aesthetics surrounding its usage. Here is where typically a shift occurs. Innovators become aware a newer technology and start the cycle over again. These innovators are typically not the innovators of the previous discipline-technology generation as most have established expertise and reputation in the status quo. However, that does not preclude a small sect of innovators that are in constant search to push the boundaries.

So where are we today in Enterprise Design (aka Enterprise Architecture not Enterprise IT Architecture)?

 

The Gartner Hype Cycle now places Cloud Technology somewhere in the Industrialization Stages 2 and 3.  However, from the artifacts and observations so far I would place Cloud maturity approaching Stage 2 as I continue to see modular componentry as the primary research thrust despite Cloud Providers, myself included**, having sold the technology as an economic benefit.  [**While at Microsoft I had been asked to research economic justification and develop a body of knowledge behind Cloud usage to create calculators and portfolio management techniques].

AI is supposedly the next big thing. From the observations and artifacts again, the discipline maturity of such is still in the Craftwork Stage where it left off in the 80s.  What has changed is the cost, performance, and availability of data for experimenting.  However, the discipline maturity is still quite low as well as the understanding of the potential hazards for a enterprise.  A decade ago program trading (AI in disguise) crashed several investment houses and several months ago a corporation had to shut down it’s AI response experiment, it had learned to become racist.  Now consider the exposure and liability a Senior Executive or CxO has with regard to such.  While AI has promise, till safety controls and the design discipline with this technology is at a Stage 3, I would suggest it be employed internally to support automation decisions with humans monitoring in failsafe positions.

Is AI a CxOs dream or an Iceberg about to be hit

 

This morning’s ponderings.  This latest iteration of AI, Machine Learning and the like seems much like the hype of the 70s AI with the exception of it operates faster.  But does that mean it operates better?  If one follow the middle page news, AI still needs a lot of work.  Yes for optimization it works, but standard OR worked too.  What is interesting are the side effects that make the news:  An AI Blog app learns to become racist, a app turns off critical systems to conserve power, and other no so minor effects that have not seen the light of day due to the sensitivity of impact. 

Bugs and side effects are normal in any new and even old technologies.  A any student of dynamic systems will tell you that side effects that you didn’t planned for are common.  These can be from your decisions to change a complex system or someone else’s decisions to do the same.  The interdependencies –some hidden from our awareness—in systems creates this complex web of unintended consequences. 

If I was a CEO would I trust AI to be making major decisions for me?  Given SOX and other standards place accountability on my shoulders I can’t push it off to “my AI system made the decision”.

This opens up two other and interdependent courses of advanced research for us at Elyon Strategies: Governance and Dynamic Systems.  While we have existing practices that utilize excellent methods for both I’m seeing both potential breakthroughs linking the two as well as a new area of concern when it comes to AI.

Personal Kanban One Month Later

Its been slightly more than a month since I had started my personal productivity experiment. This Friday afternoon I’ll be looking at all my Kanban cards (4 full pads of Post-it Notes**).  Completed, In Process, On Hold for Someone Else’s Action, and Ready Queue.  Part of that retrospective is to look for patterns and analyze where not only throughput, but effectiveness can be improved.

Just a cursory look at tasks completed provides no surprise; where I have no dependencies upon anyone on executing a task these tasks are completed quickly with a few exceptions, home projects and maintenance activities.  Where there are coordination or decision dependencies many tasks are often in the hold queue waiting for input than the level of effort to complete.  The small insight I have into this delay is due to just a few factors:

·         First the nature of opportunities is such that others need to jump on these immediately.  Whether that is to assess these or actively pursue those interruptions are necessary but still delay current activities.

·         Second, schedule integrity of all considered.  This often is caused by the prior, but often is because priority changes are not communicated until after an appointment has been missed.  When I was at Microsoft, this was a common issue along with overcommitting schedules (double or triple booking).  Typically, many had back-2-back meetings though they knew it took 10 to 15 minutes to go between meetings. The results of which was having to go over materials that were just covered 10 minutes prior.   Many times, this could have been avoided by keeping schedules on-line, ensuring that you don’t request a meeting time that was already scheduled, or sending a notification to others as soon as you know you have a conflict.

·         There are a few other reasons for such, however, what I’ve seen is the first reason dominating those dependency delays.

To address such from my end I’ve tried to keep some specific office hours, not quite free time but time where I do tasks that can be interrupted with little detriment to schedules dates.  One executive management’s end they attempt to have regular 1:1 meetings or stand-ups with teams, but again opportunities impact such.  Since I don’t have insight to management’s day-2-day activities and challenges I’m not sure of a means I could propose to help attenuate such interruptions or even if doing so would be beneficial.

The second insight which was not surprise and was pointed out in both books — Personal Kanban and Making Work Visible—was that of shifting priorities.  Here again changes in task execution priorities revolved around opportunities that arise suddenly.  I’m fortunate to be working in an area that has limited impact by such, but this can’t be avoided in a small firm.  The one saving grace to this, is that we have a highly motivated, cooperative and smart team. So, when a challenge arises it’s all hands-on deck almost before a request is made hands are raised to help.

The last insight I have before delving deeply into the statistics later today is that creating the Kanban board and pushing many items back into the ready queue limiting my work in progress has resulting in what I believe has been higher quality output and more throughput.

For my readers I would highly recommend reading and experimenting with the concepts in Personal Kanban and Making Work Visible.

Has this made a positive impact to the rest of the firm. I don’t know yet.  I plan to bring up the topic during our strategic planning session and see if others are interested in trying these concepts out, several of which require a little more coordination/commitment to our planned team stand-ups.  

**It should be noted that these cards cover tasks not the major projects these support, those some of these cards represent complete deliverables.                                       

 

Enterprise Physics -the Gravity of Business

 

I While I was preparing a presentation a few years ago for Microsoft’s TechReady 2016 I had been fairly engaged with the ideas around physics or more aptly put applying concepts of physics applied to business.  This idea intrigued me and has actually driven me and my research around creating a design system and practice for enterprises.  This vision was inspired by my mentor at IBM, John Zachman.  His articulation of using the interrogatives as a means for describing an enterprise captured my imagination and over the years has had me research to create the coordinate system behind such.
With that as a background another aspect of looking at how to apply physics to enterprise has me looking at the parallels between Quantum Mechanics and Relativity.  This week Geoffrey Moore posted a great article: A Quantum Theory of Venture Capital Valuations.  This was in addition and possibly opposition to some of the analogies he used in his book Escape Velocity which inspired the following graphic and research on innovation I was going to present at TechReady.   

The past several months I’ve been heavily investigating various physics concepts and how these apply to the world of enterprise.  Such brainstorming has led to the following graphic/concept I quickly scribbled in my notebook several months ago.

On the surface it seems readily apparent a large enterprise due to its mass warps the ecosystem it operates within and influences the behaviors of other enterprises.  This might explain why social networks are so powerful outside and inside corporations.  And why enterprises are only now catching up to such an idea.  Most corporations are still in transition from Newtonian Physics to Relativity and Quantum Mechanics.  The idea about some invisible force –social networks—effecting their organization is about as easy a concept to accept as “Dark Matter” and “Dark Energy” was several years ago in physics.

 With that little bit of insight, I’ll leave you to continue my morning R&D on how to apply such in my Design and Transformation of Enterprise workshops     

Enterprise Architecture Catalog++

Every dream about having a catalog of the complete design of your enterprise?  One that not only gives you and inventory and status of components, but also gives you the relationships between those to create higher level objects: Business and Operational Models, Capabilities, Processes.  How about linking in how these play against your business and technical strategies or assessing those strategies against your abilities to achieve?   Well the wait is almost over, because I’m almost finished building it!

Several decades ago when I met John Zachman for the first time I was impressed by his curiosity about what I was doing at Rockwell.  I was using a CAD/CAM system to design a factory.  Not only the floor plan, but the systems, applications, information flow, etc.  Several years later after John had released his brilliance regarding Enterprise Architecture I got to meet up with his again.  From our talk and his insight I’ve been working on creating a CAD for Enterprise(TM) system.  One that would enable Enterprise Architects to work with Executive Teams to design the enterprise of their dreams like designing a house.  While the UX/UI will take a while longer, the Minimal Viable Product (MVP) database which is what the application will rest upon is almost complete with a usable text-based UI.   I will be considering just a few Testers for this MVP in the next few months as well as possible partnerships with graphical design tool vendors**.  Details to follow on how-to sign up.

Name(required)

Email(required)

Website

Comment(required)

Submit

Δ

**Graphical Design Tool Vendors you can contact me directly now.