Confessions of a (post) SharePoint Architect: The self-fulfilling governance prophecy

Hi and welcome to another SharePoint (post) architect confessional post. In case you are here via the good grace of whatever Google’s search relevance algorithm feels like doing today, I need to give you a little context to this post and the larger series of which it is a part. These days, I spend a lot of time working on projects beyond the cloistered confines of SharePoint; in fact, beyond the confines of IT altogether. Apart from being a cathartic release from SharePoint work, I’ve had the privilege to work with various groups on solving some very complex problems in a collaborative fashion. As a result of these case studies, I’ve become a bit of a student of various collaborative problem solving approaches and recently released a business book on the subject called “The Heretics Guide to Best Practices” co-written with mild-mannered mega-genius Kailash Awati. Despite (or because of) the book having no absolutely SharePoint content whatsoever, it managed to win an Axiom Business Book award and I feel it’s indirectly a good SharePoint governance book in its own right.

Now, for the rest of you who have been following my epic rant thus far, you will now be familiar with the notion of Ackoff’s f-laws: “truths about organisations that we might wish to deny or ignore – simple and more reliable guides to everyday behaviour than the complex truths proposed by scientists, economists, sociologists, politicians and philosophers.” Via the f-law metaphor, you now also understand why midwives are more valuable than doctors, the word “governance” should not be defined if you actually want people to understand it and that people should not be penalised for their learning.

The next f-law that we will explore provides an explanation to why organisations so consistently and persistently apply the wrong approaches to SharePoint-type projects. IT departments have genetic predisposition to falling into this trap, as do other service delivery departments such as Finance and HR when they put in ERP systems. To explain my assertion, we are going to revisit the governance diagram that I used in the first f-law. You can see it below:

I used the above diagram to to explain f-law 1 which was “The more comprehensive the definition of governance is, the less it will be understood by all”. The above diagram serves to point out that governance is not the end in mind, but the means by which you achieve a desirable future state. Without any context to an end in mind, we have to accommodate many vague potential ends. To deal with this uncertainty, we inevitably look to definitions to provide clarity about what governance means. Unfortunately, this form of “definitionisation” tends to confuse more than clarify because it sneakily starts to drive the end, rather than be the means. This inevitably results in over-engineered, over-complicated and likely inappropriate governance approaches that do more harm than good.

It should be noted that “governance” is by no means the only word that falls into this trap. Words like quality, effective, “best practice” and even “SharePoint” should all be put in the green star above too because all of these words have no inherent meaning until they are applied to a given situation or context. This point is echoed by people like Andrew (“SharePoint by itself knows nothing”) Woodward, Dux (“SharePoint doesn’t suck – you suck.”) Sy and Ruven (“Can I use this diagram in my Information Architecture book?”) Gotz.

To that end, our next  f-law expands on this notion of means vs. ends and provides you with a practical way to assess the clarity of a SharePoint goal or outcome.

F-Law 3: The probability of SharePoint success is inversely proportional to the time taken to come up with a measurable KPI

Hmm… f-law 3 is a mouthful isn’t it. For a start I used the acronym of “KPI”, which in case you are not aware, stands for Key Performance Indicator – something that we can measure to visibly demonstrate that we have not sucked and actually achieved what we have set out to do. In essence, this f-law states that the longer it takes to determine a reasonable and measurable indicator that SharePoint has been a success, the less likely your SharePoint project is to succeed.

To demonstrate this, I am going to give you one of my patent pending techniques that is highly useful in client engagements to get people to think a little differently about their approach. Let’s reuse my “from here to there” diagram above to perform a basic experiment. Check out the project below and tell me … what project is this?

Hopefully, it did not take you long to work out that this project is the Apollo moon missions. Now, for the experimental bit. Grab a stopwatch, start the clock and answer this question:

“How do you know you have succeeded with this project?”

Once you have your answer, stop the clock and note the time. I’m willing to bet that you gave one or two answers:

• You successfully landed a person on the moon
• You got the person back to Earth again

I am also willing to bet that you worked out that answer within 2 to 15 seconds of pondering my diagram. Am I right?

Now, consider for a moment the sheer scale of of this project in terms of size, risk, innovation and level of expertise required to land a person on the moon and bring them back safely. Imagine the sheer number of projects within multiple programs of work that had to be aligned. Imagine the tens of thousands of people who directly and indirectly worked on this epic project. It is mind boggling when you think about it and it is little wonder that putting a man on the moon is regarded one of mankind’s greatest technical achievements.

And then we have SharePoint…

Now let’s contrast the moon project with another one likely to be very familiar with readers. So once again, tell me what project this is…

This one takes some people a bit longer to answer, but when I ask this in workshops and conferences I sometimes get people jokingly saying “my SharePoint project!” or “a nightmare.” So once again I want you to answer the following question:

“How do you know you have succeeded with this project?”

I bet this one has you a little more stumped and is much harder to answer than the moon example above. What is funny with this one is that, when you consider that in terms of scope and size, using SharePoint to improve collaboration is a mere pimple on the butt of sending a rocket to the moon. Yet, despite the moon example being much larger in scope, cost, degree of innovation and engineering, the success criteria is clear and unambiguous to all. People can identify what success looks like very quickly. No-one will point to Venus and say “I think that’s the moon.”  You either got there or you didn’t.

Yet, when I show a SharePoint project that is framed like the above example, people have a much (much) harder time describing what success would look like. In fact, I have asked this question many times around the world and most of the answers I am offered do not hold up to any serious form of scrutiny. Consider these common suggestions of SharePoint success and my response to them:

• “People are using it.” My response: “Yeah, but people use email and the file system now, so why are you putting SharePoint in?”
• “People are happy.” My response: “I bet if I replaced the crappy coffee with a top of the range espresso machine I could make people really happy and it’s a fraction of the cost of SharePoint.”

Sorry folks, but this isn’t good enough… in fact it’s a recipe for a situation where, in the name of “governance,” you deliver a bloated, over-engineered failure.

When problems are complicated…

My two project examples above highlight a particular characteristic of problems that is at the root of the difference between the moon and SharePoint example. Consider the following common IT projects:

• Replacing your old email system with Microsoft Exchange
• Consolidating Active Directory
• Replacing your old phone system with Voice over IP system
• Upgrading your storage area network  to new infrastructure

All of these are like the moon example. None of them are easy – in fact you need specialist expertise to get them successfully implemented. But when you put each of these in the green star of my “here to there” picture, criteria for success is fairly clear and unambiguous. For example, if email comes in and goes out of everyone’s inboxes, Exchange is a usually success. If you can pick up the phone, get a dial tone and make a call, then the VOIP upgrade has been a success.

These are all examples of complicated problems. With complicated problems, the criteria for success is clear and unambiguous and there is a strong relationship between cause and effect. You can be highly confident that doing X will lead to Y. In these sorts of problems, experts can come together and analyse the problem by breaking the problem down neatly into its parts to develop a high-confidence solution. Furthermore, there are likely to be many best practices that have emerged from years of collective wisdom of implementing solutions because of that relationship between cause and effect.

Wouldn’t it be nice if reality was always like this. Project Managers and tech people would actually get on with each other! But of course, reality paints a different picture…

When complicated approaches fail…

In a 2002 discussion paper about reform of the Canadian health system, authors Sholom Glouberman and Brenda Zimmerman make a statement that is completely applicable to how most organisations approach SharePoint:

In simple problems like cooking by following a recipe, the recipe is essential. It is often tested to assure easy replication without the need for any particular expertise. Recipes produce standardized products and the best recipes give good results every time. Complicated problems, like sending a rocket to the moon, are different. Formulae or recipes are critical and necessary to resolve them but are often not sufficient. High levels of expertise in a variety of fields are necessary for success. Sending one rocket increases assurance that the next mission will be a success. In some critical ways, rockets are similar to each other and because of this there can be a relatively high degree of certainty of outcome. Raising a child, on the other hand, is a complex problem. Here, formulae have a much more limited application. Raising one child provides experience but no assurance of success with the next. Although expertise can contribute to the process in valuable ways, it provides neither necessary nor sufficient conditions to assure success. []

In this paper we argue that health care systems are complex, and that repairing them is a complex problem. Most attempts to intervene [] treat them as if they were merely complicated. [] We argue that many of these dilemmas can be dissolved if the system is viewed as complex.

The key point in the above quote is that the tools and approaches that work well with complicated problems actually cause a lot of trouble in complex problems, where certainty of an outcome is much less clear. My point is that while the notion of using SharePoint to get from “poor collaboration” to “improved collaboration” might seem logical on the surface, it is hard to come up with any sensible criteria for success. Therefore you are setting yourself up for a fail because you have made SharePoint take on the characteristics of complex problems. Without unpacking these implicit assumptions about “Improved Collaboration,” our aspirational future state will look like the diagram below. The reality is we have many aspirational future states, all hidden beneath the seductive veneer of “improved collaboration” that in reality tells us nothing.

What blows me away is that to this day most project governance material published consistently fail to realise this core issue while trying to treat the very symptoms caused by this issue!  They provide you with the tools, means and methods to chase goals which are little better than an illusion, with no means to measure progress and therefore guide the very decisions that are made in name of governance.

Without unpacking and aligning all of these different future states above, how can any SharePoint architect be sure that they are providing the right SharePoint-based enabler? If you cannot tell me the difference made by implementing a project, how can anyone else know the difference? Even if you can, how do you know that everybody else sees it the same way as you?

Is it little wonder then, that after more than a decade of trying, SharePoint projects (complex problems) continue to go haywire? While approaches to governance force a complicated lens on a complex problem and assume the goal as stated is understood by all, governance itself will be one of the root causes of poor outcomes. Why? Because governance will require people to focus in all the areas except the one that matters. When this gap in focus manifests visibly (for example SharePoint site sprawl), governance is seen as the means to address the gap. Thus governance becomes a self-fulfilling prophecy “We will get it right this time” is the mantra, all the while, we still chase those rainbows of “improved collaboration.”

Conclusion and coming next

I do not recall where I first heard the distinction between “Complicated” and “Complex” problems because I came across it some time after I discovered the term “wicked problem.” I suspect that it was the Cynefin model that first pushed my cognitive buttons on the idea, although I distinctly recall Russell Ackoff also making the distinction between complex and complicated. Irrespective of the source, I find this a hugely valuable frame of reference to examine problems and understand why SharePoint projects are routinely tackled in an inappropriate manner. With it, I have been able to give IT departments in particular, a frame of reference to understand why they have trouble with particular kinds of projects like SharePoint.

Many people in organisations do not discern the difference between a complicated and complex problem and use the tools of the “complicated problem toolkit” to address complex problems when they are inappropriate at best and will kill your project at worst. I will expand on why this happens in the next and subsequent posts. But my key takeaway is that addressing the issue of multiple interpretations of the future is not only the key SharePoint governance challenge, it is the key challenge for any complex project.

The sorts of tools and approaches that are part of the “complex problem” utility belt are numerous and are really starting to gain traction which is great. There is plenty to read on this topic elsewhere on this blog as well as people like Andrew Woodward and Ruven Gotz. The great irony is that if you do manage align people to a shared sense of what the end in mind will look like, things that might have been seen as complex will now become complicated and the traditional tools and approaches will have efficacy because outcomes are clear and the path to get there makes more sense.

In the next post and f-law, I am going to outline another chronic issue that further explains why we get suckered into chasing false goals…

 

Thanks for reading

Paul Culmsee

www.sevensigma.com.au