Confession of a (post) SharePoint architect… What are you polishing?

Hi and welcome to the latest exciting instalment in my epic series of posts on my confessions of a post SharePoint architect. I was motivated to write this series because the mild mannered shrinking violet known as Bjorn Furuknap wrote an insightful series of articles on what it takes to be a SharePoint professional. I had always planned to write on this topic as well and opted to frame it as SharePoint “confessions” because some of my approaches do not always seem mainstream (but work!) so it feels like I am confessing my sins for using them. I chose to use the word “(post)” because SharePoint is not my fulltime gig anymore. I am very lucky to do a lot of non IT work, helping organisations deal with highly complex problems. This side of my work is where most of my insights have come from and what inspired the Heretics Guide to Best Practices book.

Thus far, we have traversed a fair bit of territory via the use of f-laws – home truths about successful SharePoint delivery that focuses on areas often overlooked for various reasons. In case this is your first visit to this series, we have covered 6 f-laws so far and I strongly suggest that you read them first…

• F-law 1: The more comprehensive the definition of governance, the less anybody will actually understand it
• F-law 2: Three is no point asking users, who don’t know what they want, to tell you what they want
• F-law 3: The chances of SharePoint success is inversely proportional to how long it takes to come up with s measurable KPI
• F-law 4: Most SharePoint objectives are platitudes: They say nothing but hide behind words
• F-law 5: Confidence is the feeling you have until you understand the problem
• F-law 6: Geeks are far less important than they think they are

In this post, we are continuing with f-law 6, focusing on aspects to SharePoint delivery where geeks have a habit of being crap…

No matter how much you polish it…

In Australia, there is an old saying, that no matter how much you try and polish a turd, it will always be a turd. In the last post, I more or less stated that some geeks have a tendency to polish turds. They do this because of a combination of an inflated view of their self-importance, mental scars from ghosts of disaster recoveries past, and a bias toward something I termed dial tone governance.

Dial tone governance refers to all of the stuff that ensures that the SharePoint platform remains pristine, consistently reliable and high performing. I noted in the previous post that this echoes what quality assurance aspires to do. This type of IT assurance for SharePoint is completely necessary, but it is definitely not sufficient. If it was, lavish praise would be heaped upon us heroic geeks for consistent fantastic SharePoint delivery.

In the last post I also channelled Neo from the Matrix and suggested that being a hero like Neo is a thankless job since, for many stakeholders, the assurance of dial tone is assumed to be there. Whether this is right or wrong is not the point, because geeks do not survive their own hypocrisy on this matter. After all, no one thanks the telephone company for providing them with dial tone when they pick up the phone to make a call – they just get pissed when it is not there.

Now while I can sympathise with unloved telephone company engineers, they actually have it easy because once they provide dial tone, their job is done. This also applies to tools like Microsoft Exchange, Virtualisation, IP networks and storage. Unfortunately with SharePoint, successful delivery is not judged on whether the level of dial tone is appropriate. At the end of the day no amount of turd polishing via awesome support, consistent process or fast response time will make a crap solution anything other than a crap solution.

So this raises a couple of questions that readers should consider:

1. Am I focusing too much (or too little) on dial tone governance?
2. What are the other governance areas that I need to focus on?

As it happens, I have some data we can use to answer them.

The hardest thing…

In 2009 I created my SharePoint Governance and Information Architecture class. The class is attended by a wide variety of roles, from BAs to PMs, CIOs as well as developers and tech dudes. It has been delivered around the world with students representing just about every industry sector (including Microsoft employees). This combination of varied audience, varied industry sectors and geographic location has provided a lot of insight, because at the start of every class, I ask students to introduce themselves, and tell the class what they feel is the hardest thing about SharePoint delivery and I dialogue map the answers.

Can you see the logic of this question? By listing all of the areas that is hard about SharePoint delivery, what should emerge are the areas we should be focusing on. Why? Well the hard bits are likely to be the areas of most risk when it comes to a failed or stressed deployment.

So let’s go through the answers given to me from a few SPGovIA classes. Maybe there are some consistent patterns that emerge. It will also be interesting to see how much of it is dial tone governance.

Brisbane 2012 and Melbourne 2010

First up, here are the answers I captured from a small class in Brisbane in 2012:

• Explaining what SharePoint is
• User uptake (“People do not like new things”)
• Managing proliferation of SharePoint sites
• Too much IT ownership (“Sick of IT people telling me that SP is the solution”)
• Users don’t know what they want
• Difficulties around SP ownership because of a lack of accountability

For me some interesting things emerge already, but before we get into detail, let’s look at a Melbourne class answering this question two years earlier and see if any consistent patterns.

• Every project is “new” (“Traditional ASP.NET web site development is ‘same old same old’)
• In SharePoint you can do things in many ways so the initial design takes longer
• The solution is never the same as the initial design and the end client may not realise this. The implication is gaps between expectation and delivery
• Stakeholders don’t know what they want (“First time around what they sign off on is not what they want “)
• Projects launched as “IT projects” with no clear deliverable and no success indicators
• Lack of visibility as to what other organisations are doing
• Determining limits and boundaries (“Doing anything ‘practically’ in SharePoint is hard”).
  • For example: We improved Ux in certain areas, but to extend to the entire feature set would take too long”
• Managing expectations around SharePoint.
  • Clients with no experience think it can do everything
  • Difficulties getting information from and translating into design, so it can be implemented
• Legacy of bad implementations makes it hard to win the business owner
• Lack of governance
  • Viral spread of unmanaged sites
  • No proper requirements of “why”
  • No-one managing it

Some analysis…

The first thing that I notice is that if you go back to the start of this post and review the six f-laws, four are clearly represented here. We have stakeholders not knowing what they want which makes design hard (f-law 2), the gap between expectation and delivery (f-law 5), the problem of SharePoint projects being done as “IT projects” (f-law 6) with “no clear deliverables and no success indicators” (f-law 3). Other themes include lack of accountability and managing viral growth of sites, but the overwhelming theme that comes through for me is that of managing user expectations and buy-in.

A telling part about what is listed is that aside from the ever present issue of managing site sprawl, not too much of it is dial tone at this point. To see if this pattern continues, let’s head to Auckland New Zealand and see if the Kiwis are any more geeky than their Australian cousins…

Auckland 2011

• Gap between expectations and reality
• Accountabilities and role clarity around delivery
• Knowledge transfer and ongoing maintenance (“Not everything is written down and when people leave, key critical information is lost. For example: Business rules set up at the start are lost over time”)
• Helping people change practices (“Getting people to use it “)
• Managing the growth over time (“the challenge of a large user base wanting to use it in different ways”)
• It’s a big, complex product
• The perception of “mystique” around SharePoint (“hard to know what not to do”)
• Seen as another “IT service”
  • product selected because it’s Microsoft
  • the people who chose the product/delivering the product are IT
• Translating the capabilities of the product to the needs of the user
  • Getting the business to understand SharePoint’s capability
  • Restrictions vs freedom
• Ramp up time: The learning curve across all roles (tech and non tech)
• The challenge of user requirements: Knowing the right questions to ask

Some more analysis…

It is clear that the themes that emerged from the two classes in Australia are also consistent here. The issue of stakeholder expectations came up straight away as well as the “IT driven project” issue (“seen as another ‘IT’ service”). Once again, the only real dial tone governance issue was the problem of managing site growth over time, but even then, it was framed more of an expectations issue (“the challenge of a large user base wanting to use it in different ways”). F-law 4 also copped a mention in terms of knowing the right questions to ask to get the right user requirements.

The additional themes that I noted from this group were:

• complexity (“It’s a big, complex product“)
• change management (“helping people change practices”)
• the high learning curve of SharePoint for users
• knowledge transfer over time the challenge of a large user base wanting to use the product in different ways.

<geek alert>Now if you are reading this and you manage complex infrastructure, let me assure you that tech people were in the classes</geek alert>. Also, since Australia and New Zealand are culturally quite similar to each other, it could be argued that we are not taking into account different cultures. So let’s find out what a 2012 class in Singapore had to say…

Singapore 2012

• Trying to deal with the sheer number of features
• “A totally different kind of concept”
  • A little knowledge can be dangerous
  • If you start with the wrong footing, you end up messing it up
• Trying to deal with “I need SharePoint”
• SharePoint for an external web site was difficult to use. Unfriendly structure for a public facing website
• Trying to get users to use it (Steep learning curve for users)
• The need for “deep discussion” to ensure SharePoint is put in for the right reasons. Without this, the result is messy, disorganised portals
• The gap between the business and IT results in a sub optimal deployment
• Demonstrating value to the business (SharePoint installed, but its potential is not being realized)
• Stakeholders not appreciating the implication of product versus platform
• You are working across the entire business (The disconnect between management/coalface)
• “Everything hurts with SharePoint”
• Facilitating the discussion at the business level is hard when your background is IT

Final Analysis

Once again the answers provided by Singapore attendees is extraordinarily consistent with the other three classes we looked at. User expectations and adoption were at the forefront, complexity was there, as was the business/IT disconnect as well as demonstrating business value. The theme of platitudes (f-law 4) and confusing the means from the ends (f-law 1) was apparent with the comment about dealing with the “I need SharePoint” issue.

I also note that the Singapore group seemed to have a greater recognition of their weaknesses – particularly with SharePoint as a “totally different type of concept” quote and last comment about difficulties facilitating discussion “when your background is IT”. I also noted one potential dial-tone comment about the difficulty of deploying SharePoint as a public facing website. Another emergent complexity related theme here is the perennial problem of SharePoint’s ample supply of features (and caveats) which risks an inappropriate up-front design decision that has negative consequences later (“Trying to deal with the sheer number of features,“ “A little knowledge can be dangerous” & “If you start with the wrong footing, you end up messing it up.”)

Finally, I particularly liked the comment about the “need for “deep discussion” to ensure SharePoint is put in for the right reasons” – that one was made by one of the Microsofties who attended the class.

Conclusions and takeaways

My own conclusion from this examination is that the responses from class attendees illustrate that dial tone governance (which is best termed as IT assurance) is necessary, but certainly not sufficient. The focus on IT assurance is a reflection of the lens that IT looks through. After all, when your performance is judged on keeping things running smoothly and reliably, it makes sense that you will focus on this.

But as illustrated by the responses, it seems that IT assurance isn’t all that hard. If it was, then why didn’t dial tone topics come up with more frequency in the responses?

So IT people, always remember f-law 1. The word ‘govern’ means to steer. We aim to steer the energy and resources available for the greatest benefit to all. Assurance on the other hand provides confidence in a product’s suitability for its intended purpose. It is a set of activities intended to ensure that customer requirements are satisfied in a systematic, reliable fashion. (I didn’t make that up by the way – that is how the ISO9000 family of standards for quality management described assurance).

The key takeaway is that to be effective and successful you actually need to apply both governance and assurance. You cannot have one without the other. Whether you have the balance right between dial tone and all the other stuff is for you to decide. So rather than focus on the stuff you already know well, perhaps it is worth asking yourself what you find hard and focus there as well.

Thanks for reading

Paul Culmsee

www.hereticsguidebooks.com